A sketch of Semitic origins, social and religious

;:|i;ii|;;ii|;ii;:ii:;;i:i;,ii|||^

,', ', ■'< • '. "m ,'( M .'. . 'iV''!'' ■"<':

<I'K'I'!'X'I',!;'1'mV'i'!''''''''''/',''c 'A' I''/,''}, ,',

BOUGHT WITH THE INCOME
FROM THE

SAGE ENDOWMENT FUND

THE GIFT OF

Menrg W. Sage

1891

4yff^^o xi/i/mz...

r

GN547 B2™" ""'«"'«V Library

oiin '^'^^ 029 872 938

Cornell University
Library

The original of this bool< is in
the Cornell University Library.

There are no known copyright restrictions in
the United States on the use of the text.

http://www.archive.org/details/cu31924029872938

A SKETCH OF SEMITIC ORIGINS

SOCIAL AND EELIGIOUS

_^T|^^^

A SKETCH

OF

SEMITIC ORIGINS

SOCIAL AND RELIGIOUS

BY

GEORGE AARON BARTON, A.M., Ph.D.

ASSOCIATE PBOFESSOE OF BIBLICAL LITERATURE AND
SEMITIC LANGUAGES IN BKYN UAWB COLLEGE

Netn got*
THE MACMILLAN COMPANY

LONDON: MACMILLAN & CO., I/td.
1902

All rights reserved

COPTBIOHT, 1902,

bt the macmillan company.

M

J. S. Cnshiog & Co, — Berwick & Smith
Norwood Maaa. U-S-A.

TO

CRAWFORD HOWELL TOY, A.M., LL.D.
DAVID GORDON LYON, Ph.D., D.D.

AND

JOSEPH HENRY THAYER, D.D., Litt.D,

^taltBsaxs in 3§artstlr Sanibtrsttj

THIS VOLUME IS INSCRIBED
IN GRATITUDE AND AFrECTION

PREFACE

The studies which have culminated in this volume
have occupied much of my attention for the past eleven
years, and have previously led to the publication of sev-
eral articles. In the autumn of 1898, while giving a
course of lectures on Semitic religion, the various parts
of the subject grouped themselves so coherently in my
mind that I could no longer doubt that these studies
had led me to the discovery of the path trodden by the
Semites in the journey from savagery to civilization, in the
course of which the most characteristic features of their
social and religious life were created. Since then the
details have been worked out with as much care as the
complex duties which attach to a very comprehensive
chair would permit, and are here submitted to scholars.

The writer is well aware that to many of his fellow
workers in Semitic studies, who have been engaged in
working different mines in our large territory, any attempt
to sketch the course of Semitic evolution will seem prema-
ture and impossible. It is the writer's conviction, never-
theless, that he has chanced upon the trail along which
the Semites dragged themselves during those weary cen-
turies when they were working their way from savagery
to civilization, and that he has had the good fortune in
some places to identify their fossil footprints and to per-
ceive the meaning of those identified in many places by
others. Here and there an identification of the exact
course of the trail is not at present possible, the luxuriant
forests, the populous cities, or the overflowing seas of
later civilizations have so buried the trail under thick
jungles, massive mounds, or strata of rock. Enough of
the trail can still be detected to render its general course

viii PREFACE

certain, aud to enable us to guess with approximate accu-
racy where its course must have lain at those points which
are hidden from view. Where it is necessary to guess at
the direction of this old Semitic pathway, I have endeav-
ored to indicate the course which it seems to me most
probable that it followed. In these instances future
investigations may show that I have not divined with
exactness all its windings and curves, but such knowledge
will be welcomed by none more gladly than by myself.

The study of primitive Semitic life necessarily brings
to view many unsavory details. Professional students
will readily understand the necessity for treating these
in the spirit in which it is done. Should this volume
chance to fall into the hands of any others, they are re-
minded that it is a study primarily not of the pure white
lily which has sprung from Semitic soil, but of the chem-
istry of that soil itself. At the conclusion of the book
the lily is not only described and appreciated, but a point
of view is gained where it can be valued the more highly
because we know the blackness of the mire from which
it springs. The Power which could bring such purity
from such unpromising antecedents impresses us anew,
and, when we reflect a little further, the wisdom of the
Providence, who prepared in such a soil the very elements
which the lily needed for its earthly nourishment, shines
out in clearer light. Such a reader is asked to judge the
book not from the first impression which its sociological
studies may make upon him, but by the vantage ground
gained at the end.

In the preparation of this volume I have been greatly
helped by my colleague, Professor Lindley M. Keasbey,
who first called my attention to the economic importance
of the palm tree, who has given me much indispensable
information in regard to sociological literature and theo-
ries, and has made many valuable criticisms and sugges-
tions concerning the sociological portion of the work.
Without his aid this portion of the book must have been

PREFACE ix

far more imperfect than it is. My thanks are also due to
my colleague, Miss Florence Bascom, Ph.D., who, in like
manner, rendered invaluable aid in those portions of the
work which touch upon geological data, and to my friend,
Professor W. Max Miiller of Philadelphia, who gener-
ously loaned me from his library books bearing on the
Hamites which were otherwise inaccessible. I am also
greatly indebted to my wife, who has drawn the maps for
this volume, carefully read all the proofs, and made many
valuable criticisms and suggestions. My obligations to
other scholars are numerous and great. An endeavor has
been made in the foot-notes to acknowledge these, but in
Fome parts of the work, as in the chapter on Yahwe, the
names of some of those whose work has indirectly con-
tributed much to my thought could not be made to appear
in a definite reference. To all such I thankfully ac-
knowledge my obligation. The articles of Thomas Tyler
in the Jewish Quarterly Review of July, 1901, and of Hans
H. Spoer in AJSL. of October, 1901, on the Tetragram-
maton reached me too late to be noticed in discussing
the origin of the name " Yahwe." I cannot see, however,
that they would have materially changed the treatment
of the subject. The references to Hilprecht's OBI. are to
the form in which the work first appeared. Those who
have the later reprint should add 214 to the number of
the page references in Part II in order to find the refer-
ences in their edition.

I cease work upon the volume, conscious of its many
imperfections, but with the hope that it may contribute a
little to the knowledge of its great theme.

Bbtn Mawb, Pa.,
November, 1901.

CONTENTS

CHAPTER I

rAsx
Thb Cradlb of the Semites 1

CHAPTER n
Pkimitivk Semitic Social Life 30

CHAPTER HI
Semitic Religious Origins 81

CHAPTER IV

Transformations among the Southern and Western

Semites 123

CHAPTER V
Transformations in Babylonia 155

CHAPTER VI
Survivals 233

CHAPTER VII

Yahwe 269

xi

xii CONTENTS

CHAPTER VIII

Brief Estimate of Semitic Social and Religious Influ-
ence ON THE non-Semitic World 309

General Index 323

Index of Scripture References 339

J

ABBEEVIATIONS USED IN THE SUBSEQUENT
PAGES

AJSL. American Journal of Semitic Languages and Literatures.

AL^2,3,4. Assyriche Lesestuke, von Friedrich Delitzsch, 1st, 2d, 3d,
and 4th editions.

BA. Beitrdge zur Assyriologie und semitischen Sprachwissenschaft,

herausgegeben von Friedrich Delitzsch und Paul Haupt.

CIS. Corpus Inscriptionum Semiticarum.

CTBM. Cuneiform Texts from Babylonian Tablets, etc., in the British
Museum, London, 1896-1901.

Assyrisches Handworteriuch, von Friedrich Delitzsch, Leipsig,
1896.

HWB.

JAOS. Journal of the American Oriental Society.

JBL. Journal of Biblical Literature.

KAT'. Keilinschriften und das alte Testament, von E. Schrader, 2d
edition.

KB. Keilinschriftliche Bibliothek, herausgegeben von E. Schrader.

OBI. The Babylonian Expedition of the University of Pennsylvania.

Series A : Cuneiform Texts. Vol. I, Old Babylonian In-
scriptions, edited by H. V. Hilprecht, Philadelphia, 1893 to
1896.

PAOS. Proceedings of the American Oriental Society.

Petermann's Mittheilungen ; i.e. Mittheilungen aus Justes Perthes geo-
graphischer Anstalt loichtige neue Erforschungen auf dem
gesammelt Gebiete der Geographic, von A. Petermann.

R. The Cuneiform Inscriptions of Western Asia, edited by Henry

Eawlinson. I H., II R., etc.. Vols. I, II, etc., of the same.

SBOT. The Sacred Books of the Old and New Testaments, edited by
Paul Haupt.

ZA. Zeitschrifte fur Assyriologie.

ZATW. Zeitschrift der alttestamentliche Wissenschaft, edited by Stade.

ZDMG. Zeitschrift der deutschen Morgenldndischen Gesellschaft.

A SKETCH OF SEMITIC ORIGINS,
SOCIAL AND RELIGIOUS

CHAPTER I

THE CRADLE OP THE SEMITES

In approaching a study of the social and religious
origins of the Semitic peoples, it is necessary, first of
all, to ask: Where did the Semitic race take its rise?
Where was it differentiated from other races, and in
what environment of climate and soil were its early insti-
tutions born ? Man, like all other creatures, is profoundly
influenced by his surroundings. The sturdy character of
the Anglo-Saxon, though for a time it may survive in the
tropics, is not created there ; nor has the careless laziness
of the negro been bred in the arctic north. To under-
stand the earliest religious conceptions of the Semitic
peoples, we must study the social organization in which
they had their birth ; and to form a correct theory of
their social organization, it is necessary to study its physi-
cal environment.

Our inquiry is, however, beset at the very threshold
with grave difficulties. The evidence with which we
have to deal is very slight, and is differently interpreted
by different scholars. The best authorities widely differ.
In recent years, four different theories as to the location
of the Semitic cradle land have been put forward, in
which Babylonia, Arabia, and North Africa are respec-
tively made the primitive home of these peoples.

1. The advocates of the Babylonian theory have been
von Kremer, Guidi, and Hommel.
B 1

SEMITIC ORIGINS

Von Kremer set forth his views in two articles pub-
lished in 1875, in Das Ausland?- He reached his results
from a comparison of the vocabularies of the different
Semitic tongues. He concluded that before the forma-
tion of the different Semitic dialects, they had a name for
the camel which appears in all of them ; whereas they had
no common names for the date-palm and its fruit or for
the ostrich. The camel the Semites knew while they were
yet one people, dwelling together; the date-palm and
ostrich they did not know. Now the region where there
is neither date-palm nor ostrich, and yet where the camel
has been known from the remotest antiquity, is the great
central tableland of Asia, near the sources of the Oxus
and the Jaxartes, the Jaihiin and Saihiin. Von Kremer
thinks the Semitic emigration from this region preceded
the Aryan or Indo-European, perhaps under pressure from
the latter race ; and he holds that the Semites first settled
in Mesopotamia and Babylonia, which he looks upon aa
the oldest Semitic centre of civilization.

Similarly, the Italian Orientalist, Ignazio Guidi, wrote
in 1879 a memoir upon the primitive seat of the Semitic
peoples, which appeared among the publications of the
Reale Academia dei Lincei.^ His line of argument and
his conclusions are similar to those of von Kremer. His
method of induction appears to have been somewhat broader
than von Kremer's, whose work seems to have been un-
known to him. He took into consideration the words in
the various Semitic languages which denote the configura-
tion of the earth's surface, the varieties of soil, the changes
of the seasons and climate, the names of minerals and ani-
mals. He concluded that Babylonia was the first centre of

1 This article was published in Das Ausland, Vol. IV, ¥os. 1 and 2,
and was entitled "Semitische Culturentlehnungen aus dem Pflanzen und
Thierreiche. "

2 The title of Guidi's paper was " Delia sede primitiva dei popoli Semi-
tic!." It is inaccessible to me ; my account of it is drawn from Wright's
Comparative Grammar of the Semitic Languages, p. 5.

THE CKADLE OF THE SEMITES

Semitic life, and that the primitive Semites in Babylonia
were immigrants from the lands south and southwest of the
Caspian Sea. This conclusion Driver, in the second edition
of his Use of the Tenses in Hebrew} was inclined to accept.
Not radically different from this is the view of Hommel,
also published in 1879. Like Guidi, he held that lower
Mesopotamia, i.e. Babylonia, and not upper Mesopotamia
on the one hand nor Arabia on the other, was the home of
the primitive Semitic people.'^ This view was accepted
by Vlock in the article " Semites," in Herzog's Real-
Encyelopedie.^ Hommel has since shifted the primitive
home to upper Mesopotamia, and now holds that it
was the home of these peoples before the separation
of the Semites from the Hamites, or, at least, from
the Egyptian branch of that stock. Egypt was, he
thinks, colonized from Babylonia, so that the civiliza-
tion of the former country was derived from that of the
latter.*

This linguistic method of investigation is, however,
precarious. As Noldeke has pointed out, the fact that
one word now denotes an object in all the Semitic lan-
guages, may be due to borrowing from one tongue by
another in remote centuries, the causes of which we can-
not now trace, while the fact that a word is not common
to all the languages of the group may not necessarily
signify that the primitive Semites were ignorant of the
object which it connotes, but may be due to the displace-

1 Cf. p. 250 n.

2 See his Die Namen der Sdugthiere lei den sudsemitischen Volkern,
Leipzig, 1879, p. 406 ff. ; and Die semitischen Volkern und Sprachen, I,
1881, p. 63.

8 For a translation of Vlook's article see Hehraica, II, p. 147 f.

« See his article " Ueher den Grad der Verwandtschaft des Altagyp-
tischen mit dem Semitischen," Beitrdge zur Assyriologie, II, p. 342 ff.
(1891-2); also "Die Identitat der altesten babylonischen und agyp-
tisohen Gottergenealogie und der babylonischen Ursprung der agyptischen
Kultur," Transactions of the International Congress of Orientalists,
London, 1892, pp. 218-244 ; and his article, " Babylonia," Hastings's
Dictionary of the Bible, 1898.

SEMITIC ORIGINS

ment of the term by another under circumstances which
now escape us.^

2. Opposed to the view that Mesopotamia is the cradle
of the Semites, is the view that Arabia was the primitive
home. This theory was defended by Sprenger in 1861 ^
and has since been reaiBrmed by him. He regards it as
an historical law that agriculturists do not become nomads,
and declares that he would as soon think that the dolphin
formerly dwelt on the height of the Alps, or the goat in
the sea, as to think that mountaineers would become
nomadic. Then, after describing the Nafud and the gen-
eral features of central Arabia, he concludes : " It is of no
importance whether the inhabitants are autochthones or
are from other neighboring tribes, the Nejd is the fastness
of the above-mentioned lands (Syria and Mesopotamia),
which has impressed its character upon the Semites."^
In like manner, in his later work, he says : " All Semites
are, according to my conviction, successive layers of
Arabs. They deposited themselves layer on layer; and
who knows, for example, how many layers had preceded
the Canaanites whom we encounter at the very beginning
of history?"*

Sayce also, in 1872,^ declared: "The Semitic traditions
all point to Arabia as the original home of the race. It
is the only part of the world which has remained exclu-

1 See Noldeke's Semitischen Sprachen, Leipzig, 1887, p. 3 ff. (2d ed.
1899), and liis article, "Semitic Languages," in the Encyclopedia Bri-
tannica, 9tli ed.

2 See his Das Lehen und Lehre des Mohammad, Berlin, 1861, I, p. 241 H. ;
also his Alte Geographie Arabiens, 1875, p. 293.

^ " Gleiohveil ob die Einwohner Autochthonen sind Oder aus andern
Gegenden stammen, das Nejd ist die Veste jener Lander, welche den
Semiten ihren Charaltter aufgedruct haben." Leben. mid Lehre des
Mohammad, Vol. I, pp. 242, 243.

* " AUe Semiten sind naoh meiner Ueberzeugung abgelagerte Araber.
Sie lagerten sioh Schiohte auf Schichte, und wer weiss, die wie vielte
Sohichte zum Beispiel die Kanaaniter, welche wir zu Anfang der Ge-
.schichte wahrnehmen, waren." Alte Geog. Arabiens, p. 293.

^ Assyrian Grammar, p. 13.

THE CRADLE OF THE SEMITES

sively Semite." The racial characteristics — intensity of
faith, ferocity, exclusiveness, imagination — can best be
explained, he thinks, by a desert origin.

Schrader, in 1873, i expressed views of the same nature.
As a result of a long examination of the religious, lin-
guistic, and historico-geographical relations of the Semitic
nations to one another, he concludes that Arabia is the
cradle of these peoples.

De Goeje also, in his academical address for 1882,^
declared himself in favor of the view that central Arabia
is the home of the Semitic race, as a whole. Like
Sprenger, he lays it down as a rule that mountaineers
never become inhabitants of the steppe and nomadic
shepherds, and so rejects the notion that the Semites can
have descended from the mountains of Arrapachitis to
become dwellers in the plains and swamps of Babylonia.
He shows, in contrast, how nomads are constantly passing
over into agriculturalists with settled habitations; how
villages and towns are gradually formed, with cultivated
lands around them; and how the space needful for the
pasture land of the nomad is gradually curtailed until,
for want of land, he is compelled to go elsewhere. So
it was, he holds, with central Arabia; and, as a result,
its nomadic population was continually overstepping
bounds in every direction and planting itself in Oman,
Yemen, Sjnria, and Babylonia. Successive layers of emi-
grants would push their predecessors farther forward until
the whole of Mesopotamia, and even portions of Africa,
shared the same fate.

Wright, whose account of De Goeje's work I have
largely reproduced, after giving a r^sum^ of this argu-
ment, observes 2 that this process has often been repeated
in historical times, in which Arabic emigration has flooded

1 See his article, " Die Abstammung der Chaldaer und die TJrsitz der
Semiten," ZDMQ., XXVII, pp. 397-424, especially p. 420 S.
^ Het Vaderland der Semitische Volken.
' Comparative Grammar of the Semitic Languages, p. 8.

SEMITIC ORIGINS

Syria and Mesopotamia. He therefore accepts the view
that Arabia is the cradle land of the Semitic race.

3. Still another theory, which is, in some respects, as
will appear later, a modification of the foregoing, is that
the earliest home of the Semites is to be found in Africa.
Thus Palgrave holds ^ that the strong racial resemblances
between the Arabs, Abyssinians, Berbers, etc., — espe-
cially the form of the jaw and the small calf of the leg, —
together with their social affinity and linguistic similarity,
lead to the view that the pure Semites of the peninsula
originally came from an African rather than an Asiatic
direction.

Similarly, Gerland reaches, on the basis of physical
resemblances, such as the formation of the skull, and on
linguistic grounds, the conclusion that all the Asiatic
Semites can be traced in their beginnings to the North
African regions. Gerland's view is in some respects
peculiar. He holds to the racial unity of the African
races, and regards the Semites as one of them. The Ham-
ites and the Semites are to him one people, and even the
Bantus are, he thinks, related to them.^

So G. Bertin advocated, in 1882, the view that the
Semites and Hamites originated together in Africa, that
the Semites crossed into Arabia, via Suez, and developed
their special racial characteristics in Arabia Petra.^

Nijldeke, too, in 1887,* accepted the same view; but he
put it forth not as a fixed theory, but as a modest hypothe-
sis. Brinton, in 1890,^ championed this hypothesis. He

1 Article "Arabia," Encyc. Brit., 9th ed.

2 See tlie exhaustive article "Ethnography," Iconographic Encyc,
Vol. I, which is a translation of the author's German work on the same
subject. He holds that sporadic traces of prognathism and woolly hair
among the Semites is an argument in favor of his view (cf. pp. 369, 370).

' Journal of the Anthropological Institute, Vol. XI, p. 431 ff.

* Die semitischen Sprache, p. 9. Also his article " Semitic Languages,"
Encyc. Brit., 0th ed.

^ See his Cradle of the Semites, Philadelphia, 1890 ; also his Baees and
Peoples, New York, 1890, p. 132.

THE CRADLE OF THE SEMITES

attempted to localize, somewhat more specifically, the
place in North Africa whence the progenitors of the
Semites migrated. He argued that popular tradition,
comparative philology, ethnology, and archaeology, all
point to "those picturesque valleys of the Atlas which
look forth to the Great Ocean and the setting sun." His
argument from popular tradition is based on a passage in
the early chapters of Genesis, and is, as will be pointed
out below, irrelevant; but his philological and ethnological
arguments are valid, as will be shown in the proper place.

Morris Jastrow also, in a paper published under the
same cover with Brinton's,^ accepts the African origin
of the Semites, although he rejects Brinton's special
locality in the African northwest as unsupported by the
evidence.

Likewise, Keane,^ who regards Mauritania as the origi-
nal home and centre of dispersion, not of the Hamites and
Semites only, but of the whole Caucasian race, naturally
holds that the Semites are of African origin. In his latest
work,^ he regards south Arabia as the earliest home of the
Semites after their migration from African soil, and there-
fore their point of departure for their several national
homes. Ripley, after reviewing the various opinions,
concludes that "the physical traits of the Arabs fully
corroborate Brinton's and Jastrow 's hypothesis of African
descent."*

This theory, that the primitive Semitic home was in
Africa, is, as the late Robinson Smith pointed out,^ not
inconsistent with the theory that Arabia was their earliest
Asiatic home, and the point from which they dispersed.
If they orginated in Africa, the arguments for the view
that the Arabian peninsula was their cradle land after

1 Cradle of the Semites.

2 Ethnology, Cambridge, 1896, p. 392.
8 Man, Fast and Present, 1899, p. 490.

* The Races of Europe, New York, 1899, p. 376.
« Wright's Comparative Semitic Orammar, p. 9n.

SEMITIC ORIGINS

their migration from the neighboring continent are, in a
good degree, reeuforced.

Lastly, Nathaniel Schmidt suggests, in a paper read at
the Congress of Religions in Paris in 1900, that the
Semites may have entered Arabia originally from Puaint,
— Abyssinia and Somali, — and that they lived in Arabia
long enough to have received their special characteristics
from its environment.

4. Another view — that Arabia was the original home
of the Hamites and Semites, and that the former migrated
thence to Africa — finds supporters among some Egyptolo-
gists. Thus, Wiedemann holds that the autochthones of
Egpyt were a race kindred to the Lybians, and that the
Egyptians of the historical period came into the country
from Arabia,! — g^^ opinion which de Morgan shares.^

Similarly, Erman has recently expressed his conviction ^
that Arabia is the home of the whole Hamito-Semitic race,
that the Egyptian, the Berber languages, and the lan-
guages of Somaliland and East Africa are Semitic in ori-
gin, though of course corrupted by admixture with African
elements. On this view, Arabia was the cradle land of
the Hamites as well as the Semites, the Hamitic migra-
tions to the westward antedating the Semitic migrations
to the eastward and northward. Erman holds that this
westward migration took place in two streams, one to
Egypt and North Africa, the other to East Africa; while
the poor region of Nubia possessed nothing to attract
Semitic settlers.

1 Cf. de Morgan's Becherches sur les origines de V Egypte,Yol. II (1897) ,
pp. 219, 223, and 228.

2 Cf. op. cit., Vol. I, p. 196 ; Vol. II, pp. 52, 53. But cf. W. Max Miiller's
review in Oriental Literaturzeitung, I, 78 ff. This theory has also been
successfully combatted by Sergi ( The Mediterranean Mace, pp. 90-100)
■who has shown that the Naqada tombs do not reveal a race different from
the later Egyptian, but simply in an earlier stage of development.

8 Cf. Erman's article "Die Flexion des agyptischen Verbums," in
Sitzungsberichte der leg. Ak. d. Wiss. su Berlin, 1900, pp. 317-353,
especially pp. 350-353.

THE CEADLE OF THE SEMITES

Is it possible, in view of such slight and contradictory
evidence, and such conflicting opinions, to find on this
matter any secure standing ground? In endeavoring to
answer this question, we may take our point of departure
from the kinship and differences between the Hamitic and
Semitic languages. These groups of languages, strikingly
different in many respects, present, notwithstanding,
some striking resemblances. Two Egyptologists, Erman
and W. Max Miiller, hold that the roots of the Hamitic,
like those of the Semitic tongues, were originally tri-
literal.i The pronoun, ordinarily the most_sui generis of
the parts of speech in a group of languages, presents, in
these groups, a similarity so striking as to point to an
original identity. The verbs in each of these groups are,
in broad outline, constructed on the same method. They
have, for example, in each group but two inflexional forms
for tenses, — -a perfect or aorist, which denotes that an
action is completed, and an imperfect or durative, which
denotes that it is incomplete. Each group treats the
weak verbs and derivatives in analogous ways. Each
group forms intensive stems by doubling a letter or redu-
plicating a root and reflexive stems in which, in each
group, the letter t is usually an important feature ; causa-
tive and reciprocal stems are also common to both groups.
Moreover, the general method of verbal inflexion — the
combination of a fragment of a personal pronoun with
the stem — is also common to both groups. The two
groups have also the same endings for gender (masc. w,
fem. f). Four or five of the numerals are identical,
and fifty, or possibly seventy-five, of the actual words of
the Old Egyptian are identical with Semitic words. ^ Both

1 Cf. Erman, op. cit., p. 350. W. Max Miiller made the same statement
at a meeting of the Oriental Cluh of Philadelphia in November, 1899.

2 Thus, in Semitic, "two" is expressed by the root Sn (Arabic tn); in
Old Egyptian, Coptic, and Tameseq, by sn; "six," in Semitic, by the
root m (contracted except in Ethiopic, as e.g. in the Heb. SS\ in
Hamitic, by sds (which appears in Tameseq), contracted in Egyptian to

10 SEMITIC ORIGINS

groups of languages form verbal nouns with a prefix m;
each regards a certain accented syllable in each word, or
group of words, as important, and each has therefore a
construct state. Each group has several consonants in
common (^Aleph, Waw, Yodh, ^ Ayin), and each writes
without expressing the vowels in written character. ^

These linguistic facts prove that the Semitic and Ham-
itic races formed one group of peoples for a considerable
period of time after the art of speech had been developed,
during which time the pronoun was fixed and the main
outlines of verbal and noun inflexion were formed; but
that they separated at a period so remote that the indi-
vidual names for objects in the two groups of languages
are, with few exceptions, absolutely diiferent.

Friedrich Miiller says concerning this: "The separa-

ss ; "seven," in North Semitic by Sb', South Semitic, s6', Egyptian, sfh;
"eight," Semitic, Smn, smn, tmn, tmK, Coptic, smn ; "nine," North
Semitic, tS", South Semitic, ts\ Tameseq, tzz. For other identical words
common to Egyptian and Semitic, see Erman, ZDMQ., Vol. XLVI,
pp. 107-126.

1 Por full proof of these statements, the reader is referred to Friedrich
MUUer's Grundriss der Sprachwissenschaft, Bd. Ill, Wien, 1884, pp. 226-
417, Erman's article " Das Verhaltness des aegyptisohen zu den semi-
tischen Sprachen," in ZDMG., Vol. XLVI (1892), pp. 93-126, and his
Aegyptische Qrammatik, Berlin, 1894, " Die Flexion des agyptischen
Verbums," in Sitzungsberichte d. Kg. Als. d. Wiss. zu Berlin, 1900,
pp. 317-353 ; Steindorf's Koptische Qrammatik, Berlin, 1894, Brugsch's
Qrammaire Hieroglyphique, Leipzig, 1872, Giovanni Collizza's Lingua
'Afar, Vienna, 1887, Belkassen ben Sedira's Langue Kabyle, Alger, 1887,
"Wright's Lectures on the Comparative Grammar of the Semitic Lan-
guages, Cambridge, 1890, and Zimmern's Vergleichende Qrammatik der
semitischen Sprachen, Berlin, 1898, passim, and esp. p. 181. Miiller
finds three groups of Hamitic languages: the Egyptian, embracing Old
Egyptian and Coptic ; the Lybian, embracing the Tameseq ; and the
Ethiopio, embracing the Bedza,"t[alla, Somali, Saho, Belin, and Chamir
tongues. Sometimes he includes, as on p. 225, the Dankali or 'Afar
language, which ought always to be included. To the Lybian group the
Kabyle and other Berber tongues should be added. TTuTler's list of
Semitic languages is not as large as it should be, but this does not afEect
his argument.

The argument for the kinship of these tongues is well stated in Crum's
article, "Egypt," in Hastings's Dictionary of the Bible.

THE CRADLE OF THE SEMITES 11

tion of the individual languages from the common origi-
nal occurred in such a way that the primitive language
was divided into two dialects, of which one might be
called Hamitic and the other Semitic. While the primi-
tive Hamitic language was, at an early time, divided into
individual tongues, probably because of the great number
of the individuals who spoke it, and the wide dispersion
of the races which used it, the primitive Semitic speech pre-
served for a long time a compact unity, probably because
of the small number of individuals speaking it, and the
narrow limits in which they lived. While, also, the primi-
tive Semitic language could develop itself uniformly with-
out foreign influx within the whole of the race which
spoke it, the Hamitic primitive language must at an early
time have broken up into a series of individual tongues,
in consequence of the separation of the peoples who spoke
it and the influx of strong foreign influences. Therefore,
it happens that to-day the unity of the Semitic languages
runs not only through the similarity of the articulation
and the grammatical foundation, but also to the identity
of roots and word-forms; while, on the other hand, the
Hamitic languages betray the fact that they belong
together merely by the similarity of their foundation and
the form of their roots, less often by the identity of the
material of the roots, and still less often by the identity of
the roots themselves.^ This opinion is borne out by an

1 See Miiller, op. cit., p. 225 : "Die Loslosung der einzelnen Sprachen
von gemeinsamen Grundstooke ging derart von sioh, dass sioh zunachst
die Grundsprache in zwei Dialekte spaltete, von denen der eine als haml-
tische, der andere als semitische Stammsprache bezeiohnet vferden kann.
Wahrend die hamitische Stammsprache weiderum friihzeitig, wahrsohein-
lich in Folge der grossen Anzahl der sie redenden Individuen und der
weiten Verbreitung der sie redenden Gesohleohter, in mehrer Dialekte,
respective Einzelspraohen sioh spaltete, bildete die semitischen Grund-
sprache, wahrsoheinlioh in Folge der geringen Zahl der sie redenden
Individuen und der besohrankten Verbreitung derselben, lange Zeit eine
geschlossene Einheit. Wahrend also die semitische Grundsprache sich
gleichmassig ohne fremde EinflUsse innerhalb des ganzen sie redenden
Stammes entwiokeln konnte, musste die hamitische Grundsprache in

12 SEMITIC ORIGINS

examination of the facts upon which it is based, and is
accepted by scholars of eminence.-^

If we accept it, we are led thereby to one of two alterna-
tives. Either the united Hamito-Semitic race lived at
some prehistoric time in western Asia, whence a large
number of them migrated to Africa, or they were all resi-
dent in northern Africa, whence the ancestors of the
Semites migrated to Asia. It will be noted that the lin-
guistic evidence, as stated by Friedrich Miiller, favors the
latter conclusion. The threads of identity which bind
the Semitic languages together are, as Miiller says, such
as to make it clear that the primitive Semitic speech was
spoken by a comparatively small number of people who
were for a long time sheltered from outside influences;
while the Hamitic languages, which are much less closely
bound together, must have been spoken by a larger and
more widely scattered body of people. Now, it is more
probable that a small number of people separated from the
main body and settled in Arabia, than that the race as a
whole originated in the latter country and the majority
migrated to Africa. While the latter supposition is not
impossible, it is far less likely to represent the true order
of events.

Gerland and others have adduced the ethnological argu-
ment in favor of the North African origin of the Semites.
It will be well, therefore, to ask what anthropology has to
tell us of the larger question of the origin of man, — or at
least of the white race, — of its primitive home, and its

Folge der Trennung der Stamme und der machtig einwirkenden frem-
den Einflusse f riihzeitig in eine Reihe von Einzelsprachen zerf alien. Daher
kommt es, dass die einheit der semitisohen Sprachen nicht nur in der
Gleichheit der Articulation und der grammatischen Anlage, sondern
auoh in die Identitat der Stamme und Wortformen zutage tritt, wahrend
dem gegentiber die hamitisohen Sprachen ihrer Zusammengehorigkeit
bios durch die Gleiohheit der Anlage und Identitat der Stoff-wurtzeln,
noch seltener in der Identitat der Stamme selbst verrathen."

1 Cf. Peschal, Haces of Men, New York, 1888, p. 493, and Eatzel,
History of Mankind, London, 1898, Vol. Ill, p. 182.

THE CRADLE OF THE SEMITES 13

differentiation from other races, in order to see whether
any confirmation of the solution of our problem to which
philology points can be found.

As to the part of the globe in which human life first
appeared, no unanimous verdict has been reached. Several
scientists assign the beginnings of humanity, with con-
siderable probability, to a continent which they believe
once occupied the site of the Indian Ocean, and which
stretched away from Borneo and the Philippines to Mada-
gascar. This view is supported by Hagckel,^ Peschel,^
Ridpath,^ and Keane.* Quatrefages, on the other hand,
holds * that man was developed from a lower order of life
in the region "bounded on the south and southwest by the
Himalayas, on the west by the Bolor Mountains, on the
northwest by the Ala-Tau, on the north by the Altai
range and its offshoots, on the east by the Kingkhan,
on the south and southeast by the Felina and Kuen-
Loun."

Gerland and Brinton hold still a different theory. They
maintain that the Mediterranean basin, including southern
Europe and northern Africa, is the part of the earth where
man first appeared.® Finally, Giddings ^ after a thorough
discussion of the evidence, concludes that "the habitat of
the homine species was probably a tropical or subtropical
zone, which reached half-way around the earth from Java
northwesterly to England." The discovery, in 1894, of
Pithecanthropus erectus, a kind of missing link, in Java,

> History of Creation, New York, 1884, Vol. II, p. 326.

2 The Races of Men, New York, 1888, p. 32.

' Great Baces of Mankind, Cincinnati, 1893, Vol. I, pp. 173-182.

* Ethnology, Cambridge, 1896, p. 229 ; cf. also Suess, Anlitz der Erde,
Leipzig, 1883, Vol. I, p. 535. ~~" "

5 The Human Species, New York, 1890, pp. 176-177.

* Inconographic Encyclopxdia, Vol. I, p. 29, gives Garland's view.
Brinton's is found in his Baces and Peoples, Neio York, 1890, pp. 82-94.
Gerland is not so specific as Brinton. He places the cradle of the race in
Europe, but a Europe differing greatly from the present continent.

■^ Principles of Sociology, The Macmillan Co., 1896, p. 219.

U SEMITIC ORIGINS

would lend some plausibility to the first or the last of these

views.

Fortunately our subject does not compel us to decid©
between these contending scientists. It is becoming clear
that the beginning of the human species is so remote that
man was distributed at an early date over practically the
whole earth, and that the processes of race formation have
been so gradual and so long that it is as unnecessary
as it is impossible to connect the cradle of the Semites
with the birthplace of the race.

Croll,^ by showing that the glacial epochs have been
caused by variations in the eccentricity of the earth's
orbit, which, from astronomical data, can be computed in
terms of years, has given us some idea how long ago those
men lived whose remains go back to the glacial and to
earlier periods. This method of investigation has been
widely accepted by scientists, many of whom have applied
the results in ways of their own. While, therefore, their
estimates of the antiquity of man do not agree, they are all
sufficiently large to remove the beginnings of the species
far from the beginnings of any of the races which now
exist. Thus, Quatrefages^ supposes that man goes back
to Miocene times, a geologic epoch which CroU* believes
ended about 720,000 years ago. Fiske,^ on the basis of
the discovery of the fossil remains of man by Ribeiro and
Whitney in the Pliocene rocks of Portugal and California,

1 Cf. Giddings, op. cit., p. 217.

^ See his Climate and Time in their Geological Belations, London,
1890, 4th ed. This work is a collection of papers, many of which had
appeared earlier in the Philosophical Magazine and other journals during
the ten or twelve years prior to the first edition of the book.

• The Human Species, p. 152.

* Climate and Time, p. .359.

' Excursions of an Evolutionist, Boston, 1890, pp. 36, 75-77, and 148.
Recent investigation tends to undermine the correctness of Whitney's
inferences ; of. William H. Holmes in American Anthropologist, new
series. Vol. I, p. 107 ff. Payne, however, holds (History of the New World
Called America, Vol. II, p. 64 ff.) that by the glacial epoch man wa»
present in both the old and the new worlds.

THE CRADLK OF THE SEMITES 15

is led, by CroU's calculations, to believe that man was as
widely scattered over the earth as the distance between
California and Portugal as long as 400,000 years ago.
Keane^ estimates the age of man as from 240,000 to
1,000,000 years, — a latitude rather startling, — Ridpath^
at 200,000 years, Lubbock* at hundreds of thousands of
years ; while Gerland * thinks that the facts of ethnology
point to at least 240,000 years.

Some have objected that man could not have lived so
long upon the earth without undergoing great physical
modifications. Since Miocene times natural selection has
transformed all the animals, and why should it not have
transformed man, too? Yet the oldest fossil remains are
almost identical in form with the existing races of men.
This difficulty disappears when we remember that, as
Wallace has pointed out,^ when once man's mind was
developed, natural selection would cease to act upon his
body, and would expend itself in expanding his mind.
The approach of a glacial epoch instead of rendering him
extinct, or developing a hairy covering for his body, would
sharpen his wits to enable him to provide food and cloth-
ing for himself. During the time of his existence, there-
fore, every other form of physical life may have been
entirely transformed by those laws of natural selection
which have ripened his mental power.

It is clear, from the foregoing summary of opinion, that
we may, with perfect propriety for the purpose of the
present inquiry, dismiss this larger subject of the origin
of man and confine ourselves to the investigation of a
smaller part of the question.

If we turn our attention to the origin of the white races,
the problem appears about as difficult for one not a profes-

1 Ethnolnffy, pp. 56-70.

2 Great Races of Mankind, Vol. I, p. 150.

» Prehistoric Times, 5th ed., London, 1890, pp. 383-425.

* leonographic Encyclopcedia, Vol. I, p. 28.

* Natural Selection and Tropical Nature, The Macmillan Co., 1891,
p. 180.

16 SEMITIC ORIGINS

sional anthropologist to handle, so little agreement has
been reached by those who have given the subject special
attention. Thus, Hseckel,! Peschel,^ Brinton,^ Keane,*
and Seigi ° regard the Mediterranean basin as the cradle of
the white races, while Ripley ® gives convincing evidence
that there is no white race, but that at least three distinct
white races are traceable in Europe : the Teutonic, or Homo
Europseus, most distinctly preserved in Scandinavia; the
Alpine (for which he rejects the name Celtic), most clearly
preserved among the lower classes of Austria and Bavaria,
and clearly of the same type as the prehistoric lake-
dwellers of Switzerland;^ and the Mediterranean race,
most clearly defined in Corsica, Sardinia, and the Iberian
peninsula, and clearly identical with the Berber race.^
Ripley recognizes, as the others do, that the race of the
south of Europe is identical with that of North Africa,
and that it is a white race. These are, after all, the only
points which affect our problem. Sergi holds that the
substratum of all the population of Europe is composed of
this Mediterranean, or Eurafric race.®

Gerland introduces a difficulty into the problem, how-
ever, by denying that the Semites belong to the white
race, and would connect them with the black. Thus he
says : " The Arabic-Africans are one ethnological race, or
division of mankind. . . . Kai-Koin, Bantu, and Negro
tribes exhibit the same physical characteristics as the
Semites ; and their languages show the path by which the
Semitic reached its goal."^** This opinion does not seem

^ History of Creation, Vol. 11, p. 321.
2 Races of Men, pp. 480-518.
8 Maces and Peoples, pp. 97-139.
* Ethnology, ch. xiy.
' Mediterranean Mace, New York, 1901.
<■ ITie Maces of Europe, ch. vi, especially pp. 121-130.
' Cf. The Lake Dwellings of Europe, by Robert C. Monro, London,
Paris, Melbourne, 1890.

» See op. cit., pp. 247 and 276.

' The Mediterranean Mace, New Yoi-k, Scribners, 1901.

i» Iconographic Encyclopedia, Vol. I, pp. 369, 370.

THE CRADLE OF THE SEMITES 17

to rest on a sufficiently secure basis. The occasional
prognathism and woolly hair observed among the modern
Arabs, on which he relies,^ is much more likely due to
mixture brought about in recent times by the Arabic slave
raids into Africa, which have been going on for centuries.
Travellers in Arabia testify that, in consequence of this,
amalgamation of races is still going on there. ^

The Bantu languages, as described by Miiller,^ exhibit,
so far as I can see, no real kinship to the Semitic tongues.
The pronouns are not similar to the Semitic, as is the case
with the Hamitic pronouns; the tenses of the verbs are
more like those of the Aryan than those of the Semitic
tongues, so that the similarity seems to be reduced to the
fact that the Bantu languages have causative, reflexive,
reciprocal, and causative-reflexive stems. These are, how-
ever, formed, for the most part, by afformatives instead of
preformatives, as in Semitic, and cannot be held to prove
kinship. It seems safe, therefore, to follow the prevailing
opinion and class the Semites with the white races. Sergi,
moreover, has pointed out that many of the Hamites in
North Africa are to-day blonds, and that they are probably
native there. The pigmentation of the skin is, he believes,
due to the altitude of the country in which a people lives.*
If this be true, the question of complexion need not
seriously trouble us.

If now, with Ripley, we conclude that the Alpine race
is identical with the race of the prehistoric lake-dwellers,
we shall also look for the perpetuation of very ancient
racial types in the Mediterranean race. The Mediter-
ranean basin has at different times undergone such changes
of level that its north and south shores have been united
at Gibraltar, or by way of Italy, Sicily, and Tunis, or in

1 See above, p. 6 note 2.

» Cf. Palgrave's Central and Eastern Arabia, Vol. I, p. 452, Vol. II,
pp.242, 272, and 302; also Doughty's Travels in Arabia Deserta, Cam-
bridge, 1889, Vol. I, p. 553, Vol. II, pp. 80, 171, and 337.

" Cf . Gfrundriss der Sprachivissenschaft, Vol. I, Pt. II, pp. 238 ff.

* T%e Mediterranean Mace, pp. 59-75.
c

18 SEMITIC OKIGINS

both places.^ This fact accounts for the unity of the
fauna and flora on the two sides of the sea. At the same
time the Sahara desert was, at least in part, submerged,
so that the northern part of Africa was separated from the
southern and, through long lapse of time, the life of the
two parts of the continent became distinct. This condi-
tion was probably terminated by the beginning of the last
glacial epoch. It is probable that the Mediterranean race
was, in this far-off time and under these conditions,
developed.

Boyd Dawkins,^ who is followed by Fiske,^ tells the
story of the succession of the races of men in Europe in a
most attractive way. If we could follow him we might
enter upon some very pleasing speculations. His results
are not admitted by the great number of anthropologists,
however, and seem to be based on insufficient data. We
must therefore be content to follow a more sober path.

If we take the more sober statement of Ripley and Sergi,
that the race still distinctly marked in Corsica, Sardinia,
and the Spanish peninsula is identical with the Berber
race of North Africa, we have the clew for which we have
been seeking. The late Count von der Gabelenz,* following
the lead of certain ethnologists who thought they detected
racial affinities between the Basques and Berbers, endeav-
ored to go further, and to show that the tongue of this
little people of the Pyrenees is kindred to that of the
Berbers of North Africa. If this could be established,^

1 Cf. Suess's Anlitz der Erde, Leipzig und Prag, 1883, Vol. I, p. 771 ;
Neumahr's Erdgesckichte, Leipzig und Wien, 1890, Vol. II, p. 698 ; and
"Wallace's Geographical Distribution of Animals, The Maomillan Co.,
1876, Vol. I, pp. 113-115, and 201 ff.

^ Cf . his Man in Britain and his Place in the Tertiary Period, London,
1880, pp. 161-172, and oh. vii.

' Cf. his Excursions of an Evolutionist, pp. 39, 44-46.

* See Die Verwandtschafl der Baskischcn mit den Berber-sprachen
Nord-Afrikas, Brunswick, 1894. He published a previous essay on the sub-
ject in the Sitzungsberichte der Ak. d. Wiss. zu Berlin, 1893, pp. 593-613.

* See Brinton's Races and Peoples, p. 112, and Keane's Ethnology,
p. 378.

THE CKADLE OF THE SEMITES 19

the threads which connect the Berbers with the races of
the Iberian peninsula would be greatly strengthened; but
his effort cannot be pronounced a success. He admits that
the languages differ in structure of speech, in gender, and
in most of the formatives ; but urges that they have cer-
tain analogous laws of phonetic change, and that there is
a resemblance in a few culture words, such as the names
of animals and of articles of dress. These names, how-
ever, afford no basis of argument whatever, as they may all
have been borrowed during the Arabic-Berber occupation.
Stumme ^ has pointed out that they seem quite as much
Arabic as Berber — a fact which seems to make the hypothe-
sis just advanced the more probable. The labor of von
der Gabelenz seems to have been suggested by a mistake,
for Ripley now comes forward and proves that there is no
pure Basque type.^ Without the aid of philological argu-
ments, however, we may rely upon the fact, in which all
anthropologists agree, that the race on the two sides of
the Mediterranean is identical.

We noted above that the geologic changes which sepa-
rated North Africa from Europe occurred by the beginning
of the last glacial epoch. CroU ^ calculates that that epoch
began about 240,000 and ended about 80,000 years ago.
The free interchange by land necessary to make this iden-
tity of race must, therefore, have occurred before this
remote period, and must have been going on before that
for a time sufficiently long to fix a racial type so constant
that it still persists on both sides of the Mediterranean.
During the millenniums which have elapsed since this
epoch this race has persisted in these regions, has absorbed
all foreign elements which have been injected into it, and
has maintained its identity in the face of everything.

1 Cf. Literarisches Centralblatt for 1895, p. 581. For other criti-
cisms of von der Gabelenz's work, see Berlin philologisches Wochen-
schrift, Vol. XXV, p. 784, The Academy, Vol. XLIV, p. 93, and Science,
Vol. XXII, p. 77.

' Op. cit., ch. viii.

s Climate and Time, chs. xix and xx, especially pp. 328 and 342.

20 SEMITIC ORIGINS

So far as I know, the general unity of the Hamito-
Semitic stock is not seriously questioned.^ We may grant
that, as they moved out from their original home, they
may have imposed their languages upon foreign tribes,
and thus have become, in course of time, somewhat modi-
tied at the extremities ; ^ nevertheless, the general racial
type is well marked, showing that, on the whole, they are
rightly classed together. Among the Hamitic peoples we
may expect to find traces of racial mixture in Somaliland,
among the Dankils, and the people of that region; but
for the rest of the Hamitic stock a purer racial type appears
to exist, and the kinship of the Hamitic tongues, which
links the Berber languages to the ancient Egyptian and
to those of East Africa, and through these to the Semites,
proves that there exists either a real kinship, or that
through conquest one race has imposed its language upon
the rest. The latter alternative is ruled out, it seems to
me, notwithstanding the views of Erman'^ and others, by
the consideration that the records of a part of the Hamitic
race go back to the very dawn of history, and that in these
records from Egypt we have no trace of such an extensive
conquest as would be required to account for this unity of
language. Such a conquest would have to be early to
allow time for the tongues to develop their striking differ-
ences. At such an early time we have no record of a con-
quest of large parts of North Africa, and the conditions
for it did not exist. Further, as Sergi has pointed out,*
the fact that the Berbers developed an independent system

1 Sergi seems to imply that they are distinct, for example in his discus-
sion of the Hittites and Phcenioians (Mediterranean Race, pp. 150-163),
but he does not enter fully into the subject.

2 See Kipley's Baces of Europe, p. 376.

' See Sitzungsberichte d. kgl. Ak. d. Wiss. zu Berlin, 1900, pp. 350-
353. Cf. also de Morgan and Wiedemann, in de Morgan's Becherches sur
les Origines de VEgypte,Yo\. I, p. 196. Deniker accepts de Morgan's
views, but is in doubt whether the Hamitic peoples originated in Asia or
Europe. Cf. his Baces of Man, London, 1900, pp. 426, 428.

* Mediterranean Race, p. 50 ff.

THE CRADLE OF THE SEMITES 21

of writing, wholly uninfluenced by hieroglyphic Egpytian,
proves that no such conquest can ever have taken place.
We must therefore conclude that the real explanation is
in the fact of kinship.

If, then, the Berbers are a real part of the Hamitic race, i
and at the same time are a part of that Mediterranean race 1
which has been resident in this region since the last glacial
epoch, we have at last some secure ground on which to
tread. It becomes clear that the southern shore of thel
Mediterranean was the original home of the Hamito- '
Semitic race ; ^ that at some time since the glacial period,
but after the germ of languages still spoken had begun to
develop, the Semites were separated from their Hamitic i
brethi'en, and in their migrations ultimately reached;
Arabia, and that the more numerous Hamites gradually
spread themselves over the northern part of Africa. Eth-
nology thus confirms the conclusion to which the linguistic
phenomena of the two families of languages — the Semitic
and Hamitic — had led us; viz. that the Semites migrated
from Africa, and not the Hamites from Asia. How long
ago these movements began we cannot tell. If the glacial
epoch ended 80,000 years ago, who knows how much of
the period which has elapsed since then may not have
been occupied by the various movements which have
transformed the Hamito-Semitic family into the nations
known to history?

In this connection we may note that the arguments of
Hommel in favor of the Babylonian origin of the Egyptian
civilization, 2 even if they were much stronger than they
are, would have no bearing on the point in question ; for
if it could be proven that the oldest monuments of Egyp-

1 So Paulitschke holds that the Hamites were autochthenes of the
northern coast of the African continent, Beitrage zur Etlinographie und
Anthropologic der Somdl Galla und Horari, 2ded., Leipzig, 1888, p. 7.

2 See his article, " Ueber den Grad der Verwandtschaft des Altagypti-
schen mit dem Semitischen," in Beitrage zur Assyriologie, Vol. II, pp. 342-
358, and his article, " Babylonische Ursprung der agyptischen Kultur,"
in Trans. Inter. Cong. Orient., London, 1892, pp. 218-257.

22 SEMITIC ORIGINS

tian culture were inspired by influences from Mesopotamia,
it would only demonstrate that there was a western move-
ment of migration from the Tigris-Euphrates valley thou-
sands of years later than the time of which we are speaking.
That there have been many migrations of Semites in vari-
ous directions, no one acquainted with early history will
deny. Should Hommel's contention be granted, then
(and I am by no means convinced that it should be), it
has no more bearing on the movement in the opposite
direction, in the remote period under discussion, than the
Mohammedan conquest of Egypt has.

The arguments of von Kremer, Guidi, and Hommel,
referred to above, in favor of regarding Babylonia as the
earliest centre of Semitic culture — the cradle of the
Semitic race into which it was put soon after its birth
in the high regions east of the Caspian Sea — cannot be
regarded as credible in opposition to the Afro-Arabic
origin.

Their argument is met by the following objections:
(1) It is based on linguistic data, which Bertin^ and
Noldeke^ have shown to be precarious. It is, as Bertin
points out, precisely the objects which are most common
for which most synonyms exist in a language, some of
which would survive in one of its derived tongues and
others in another. (2) It places the primitive Semitic
home in a region where, at the dawn of history, the
Semites were in conflict with other races, from one of
which many scholars hold that the resident Semites bor-
rowed much of their civilization. These races so pro-
foundly modified the Semitic spoken in the region that
it has suffered more deterioration than any other Semitic
tongue. Surely it will take more than linguistic evidence
to convince us that the primitive home, in which the
Semites developed those special characteristics which so
strikingly differentiate them from other races, was in such

* See Journal of the Anthropological Institute, Vol. XI, p. 426.
^ Die semitischen Sprachen, p. 3 ft.

THE CRADLE OF THE SEMITES 23

a region. (3) This theory would compel us to believe
that the Semites migrated from the fertile plains of Baby-
lonia to the wastes of Arabia ; and it offers no sufficient
motive for such migration, although it necessarily puts
the entrance into Arabia at a comparatively late date, and
has to face the fact that in historic times the movement
has been all the other way. (4) It in no way satisfac-
torily accounts for the connection between the Semitic
and Hamitic stock, and leaves utterly unexplained the
fact that the Hamitic languages are less closely related to
one another than the Semitic, though all the tongues of
the two groups are kindred. (5) The anthropological
argument, as we have seen, is against it.

Of the four classes of arguments by which Brinton ^ sup-
ports his theory of the North African origin of the Semites,
two — the traditional and the archaeological — may be
dismissed at once. The Biblical traditions embodied in
Genesis ii and iii, even if the word qedem could be trans-
lated " eastward," as he supposes,^ cannot possibly refer to
events so far in the past as the original movement of the
Semitic from the Hamitic races. This event must have
been separated from the Biblical writer by at least several
thousand years — a period through which the memory of
no uncivilized people carries a reliable tradition. The
memory of the Biblical tradition may go back to Baby-
lonia, and may, as we shall see in a subsequent chapter,
refer to practices born in Arabia and perpetuated in Baby-
lonia and elsewhere; but it certainly refers to far later
events in the history of the Semites than their primal
migration. His archaeological argument from the method
of using the bow known as the "arrow-release"^ is, for
similar reasons, of little if any significance. It is a
practice found in many parts of the world.

The linguistic and ethnological arguments on which

1 The Cradle of the Semites, Philadelphia, 1890.
" Op. cit., pp. 5, 6.

2 Op. cit., p. 11. Cf. Jastrow's criticism of it, ibid., p. 23.

24 SEMITIC ORIGINS

others had laid stress before him are, as we have seen,
valid, and in the absence of weighty reasons for any other
view must be considered convincing. We therefore hold
that North Africa was the home of the Hamito-Semitic
stock.i We cannot, so far as I can see, with our present
knowledge, settle upon any special locality there and
claim that this rather than another was the place where
this race first appeared.^ Its habitat must have included
the Mediterranean coast lands, from whence it could pass
into southern Europe.

As has been already pointed out, the theory that Africa
was the primitive home of the Hamito-Semitic race in no
way conflicts with the theory of Sprenger, Sayce, De
Goeje, and Wright, that Arabia was the specific home of
the Semites, — the country where their peculiar character-
istics were developed and the centre from whence the
Semitic nations radiated to other lands. All the argu-
ments they have urged seem to me valid, so that, as the
late Robertson Smith perceived, though the Semites came
from Africa, Arabia is the centre from which they spread.*

It will be remembered that Friedrich Miiller, in the
passage quoted from him above,* suggested that the
greater unity which the Semitic languages present, when
compared with the Hamitic, is probably due to the fact
that the primitive Semitic tongue was spoken by a smaller
number of people, who lived in a more confined area than
was the case with the primitive Hamitic. This sugges-
tion grows naturally out of the nature of the two groups
of languages, and commends itself as true. It harmonizes

1 So also Ratzel. See his History of Mankind, Vol. Ill, p. 181 ; "The
Hamites are the aborigines of Africa."

" Brinton holds that this race originated in the vicinity of the Atlas
Mountains, while Sergi formerly held, and still inclines to believe, that
East Africa, in the neighborhood of Somaliland, is their primitive home.
Cf. Mediterranean Bace, pp. 42, 43, 70.

8 Cf . Wright's Comparative Orammar of the Semitic Languages, p. 9 n.
Cf. above, p. 7, n. 6.

*p. 10 ff.

THE CRADLE OF THE SEMITES 25

also with the view that the Semites migrated from Africa
(since the emigrants would naturally be few in comparison
with those who would remain), and that within the com-
paratively narrow confines of central Arabia they long
lived in close contact with one another and separated
from the rest of the world.

One comes occasionally across a writer who holds that
the pure Arabs entered Arabia from Abyssinia.^ If this
means that they entered Arabia as Hamites, and were
afterward differentiated from the Hamitic peoples, it may
possibly be true; but if it means that Abyssinia is the
cradle of the Semitic peoples and Arabia only its centre of
distribution, the view, not to mention a social argument
which will appear in a future chapter, is, for the follow-
ing reasons, untenable : (1) The physical characteristics of
Abyssinia'* and Somaliland are not such as to have iso-
lated the Semites for a long period from foreign influ-
ences in such a way as to produce the linguistic results
which appear among them; and (2) Hamitic tribes (the
Afars or Dankils, the Gallas, and Somalis) now lie to the
south and southeast of the Semitic inhabitants of Abys-
sinia, so that the Semitic population, driven in like a
wedge, separate these Hamites from the Hamites of
Egypt.* It is clear, therefore, that the Abyssinian
Semites are late intruders in this region. The Geez

iCf. the article "Arabia" in Encyc. Brit., 9th ed., and Schmidt's
paper read before the Congress of Religions in Paris.

2 Cf . The Sacred City of the Ethiopians, being a Becord of Travel and
Research in Abyssinia in 1893, by J. Theodore Bent, London, 1893, in
which are many descriptions of the country. Cf. also Ratzel's History
of Mankind, Vol. Ill, p. 222 ff. ; and Jules Borelli, Ethiopie Meridionale,
Paris, 1890, passim ; also Modern Abyssinia, by Augustus B. Wylde,
London, 1901, ch. xx.

s Brinton's Baces and Peoples, p. LSI; Gerland, Iconographio Encyclo-
pcedia. Vol. I, p. 352 ; and Ridpath's Great Baces of Mankind, Vol. Ill,
pp. 469-472. The objection of Erman {Sitzsungsberichte der Kgl. Ak.
d. Wiss. zu Berlin, 1900, p. 352), following Lepsius, that the Nubians are
non-Hamitic and also intervene between the Hamitic nations may be
obviated by one of two considerations : either the Nubians may have

2B SEMITIC ORIGINS

civilization, especially in its ancient form,i has so many-
characteristics in common with that of south Arabia, —
characteristics, too, which appear more at home, as we
shall show below, in Arabia, — that there can be no doubt
but that it is the result of a westward Semitic movement
from that country across the straits of Bab-el-Mandeb,
and not the unmoved remnant of the primitive Semitic
stock.

If now it be asked what can have induced the Semites
to enter so sterile a country as Arabia, we can only answer
by conjectures. It must be noted that their migration
probably occurred so long ago that the Nile valley had
not begun to be cultivated, so that they were not tempted
on account of the fertility of the country to stop there.
This we infer from the fact that the language of the
earliest monuments of Egypt is so different from the
Semitic. The Hamitic Egyptians must have come into
the Nile valley and developed their civilization long after
the Semites had passed on to Arabia. ^ It is hardly prob-
able that at that remote period Arabia was a less deso-
late country than now. Wallace believed that the large
treeless tracts of desert in Eastern Africa and Western Asia
were once covered with aboriginal forests, which were
destroyed by the abundance of camels and goats, — animals
which are exceedingly destructive of a woody vegetation,
— and that the loss became permanent on account of the
absence of irrigation. ^ If Arabia was ever covered with

entered this region after the Hamites of East Africa had gone thither,
or their country may have been too poor to attract the Hamites, so that
they passed them by. As Erman confesses, this is tlie real explanation.

1 See below, ch. iv.

^ Of course it is not impossible, as Erman {Sitzungsherichte der Ak.
d. Wiss. su Berlin, 1900, p. .351 ff.) and Wiedemann (in de Morgan's
Becherches sur les Origines de I'Egijpte, "Vol. II, pp. 219-228) believe,
that a later wave of Semitic migration from Arabia may have formed
one element in the formation of the old Egyptian civilization. It was
probably not an important element. Cf. Sergi, Mediterranean Bace,
pp. 90-100.

^ Geographical Distribution of Animals, Vol. I, p. 200.

THE CRADLE OF THE SEMITES 27

forests, it must have been many thousands of years ago;
and it is unlikely that the Semitic migration can be placed
far enough back so that such an inducement can be thought
to have led its hordes hither. As our investigation pro-
ceeds, however, the reasons which led to this migration
will become clear. ^

How long ago the Semites entered Arabia we can now
only guess. It must have been, as we have already noted,
after the main skeleton of Hamito-Semitic verbal forma-
tion was formed and several of their numerals developed.
On the other hand, it must have been some considerable
time since the change in the contour of the continents
had separated those members of the Mediterranean race
resident in Europe from those resident in Africa. We
infer this because, while the Berbers are connected with
the other Hamitic races in language, there is no linguistic
connection between them and any member of the Mediter-
ranean race resident in Europe, either in ancient or modern
times, except those Arabs and Berbers who migrated into
Spain in the early days of Islam. Eighty thousand years
ago (or should we say 240, 000? 2) no fixed language
(i.e. a language sufficiently fixed to survive) had been
developed by these Mediterranean peoples. All this
might be true, and yet the two great families of Hamito-
Semitic speech might have been outlined as much as they
were when the Semites branched off, if that event be
placed at 20,000, 30,000, or even 50,000 years ago. Any
one of these periods would be sufficiently long, so that
Arabia may have been a more fertile country then than
now, and so that its present conditions may afterward
have supervened, and still have occurred long enough ago
to allow them thousands of years in which to indelibly
impress on the Semite his social organization and religion.
We shall, however, see in the next chapter some reason
for holding that the conditions of Arabia were what they
are now when the Semites entered the country.

1 See below, oh. iii. ^ See above, p. 19.

28 SEMITIC ORIGINS

However long ago this migration may have occurred, —
and any statement in years is a mere guess, — the peculiar
conditions of life which the Arabian deserts and oases
have presented for millenniums are the matrix in which
Semitic character, as it is known to us, was born. It is
a land of barren and volcanic mountains,^ of broad stretches
of dry, waste, unproductive soil,^ and wide areas of shift-
ing sand, interrupted by an occasional oasis, — a land
where, for the most part, water is difficult to obtain, where
famine is always imminent, where hunger, thirst, heat,
and exposure are the constant experience of the inhabi-
tants. The Bedawi are always underfed, they suffer
constantly from hunger and thirst, and their bodies thus
weakened fall an easy prey to disease;^ they range the
silent desert, almost devoid of life, where the sun is all
powerful by day and the stars exceedingly brilliant by
night. This environment begets in them intensity of
faith of a certain kind, ferocity, exclusiveness, and
imagination. These are all Semitic characteristics wher-
ever we find the Semites ; and there can be little doubt
but that this is the land in which these traits were in-
grained in the race. Here, too, the Arabic language, pre-
served in its purity by the barriers which nature interposed
against foreign influences, though it is by no means iden-
tical with the primitive Semitic language, has preserved
more characteristics of that primitive speech than any
other Semitic tongue.*

We conclude, then, that we must hold to the Arabic
origin of the Semites. Taking Arabia as the Semitic
cradle land, the course of distribution of the Semitic
nations over the lands occupied by them during the his-

1 Cf. Doughty's Travels in Arabia Deserta, Vol. I, chs. xiii-xvi.

2 Cf. Doughty, op. cit.. Vol. I, p. 56.

'See the books on Arabian travel generally; e.g. Doughty, op. cit.,
Vol. I, p. 244.

« Cf. Schrader in ZDMa., Vol. XXVII, p. 417 ; Wright, Comparative
Cframmar of the Semitic Languages, p. 8 ; and Vlock in Hebraica, Vol.
II, p. 149.

THE CRADLE OF THE SEMITES 29

torical period would be that described by Schrader ^ and
Wright ^ on the basis of the relative divergence of the lan-
guages from the primitive type. The northern Semites
— the Babylonians, Aramaeans, and Canaanites — first
parted from their brethren in the south and settled in
Babylonia and the neighboring regions, where they lived
together for a long period. The Aramaeans were the first
to separate from the main body of emigrants ; at a consid-
erably later period the Canaanites, and, last of all, the
Assyrians. At the same time an emigration went on in
a southern direction. Parting from the main body in
central Arabia, these emigrants settled on or near the
southern coast of the peninsula, whence a band of them
subsequently crossed into Africa and pitched in Abyssinia.
These movements must each be considered as processes
going on for a considerable period,^ and, in some cases, as
in Mesopotamia and Palestine, subject to a considerable
mixture not only of foreigners, but from Arabia directly.

1 ZDMG., Vol. XXVII, p. 421 fi.

2 Op. at., p. 9.

s Cf. Robertson Smith's Kinship and Marriage in Early Arabia,
Cambridge, 1885, p. 244.

CHAPTER II

PKIMITrVE SEMITIC SOCIAL LIFB

In primitive life the form of the clan depends upon
economic conditions. My friend and colleague, Professor
L. M. Keasbey, who is engaged upon a work on the eco-
nomic origins of society, would sketch some of the differ-
ences as follows : In protected spots where the beginnings
of agriculture or arboriculture are possible, communities
are formed of women and the weaker men. The stronger
men are drawn away by more hazardous enterprises.
Polyandry of the Nair type may prevail, and descent
will be reckoned in the female line. This gives us the
"communal clan."

Where men organize for hazardous enterprises, such as
conquest, conducting a caravan, or hunting the buffalo,
the " republican clan " is formed. The success of the
enterprise requires the most skilful leader, who must,
accordingly, be chosen for his personal qualities. A few
hardy women are taken into these clans, and polyandry
of the Thibetan type, or communal marriage, may result.
Descent is here counted through the father.

Where pastoral life is possible, the care of the flocks
leads to the formation of the "patriarchal clan." In this
polygamy may prevail, and descent is reckoned through
the father.

The late W. Robertson Smith, in beginning his discus-
sion of the relations of gods and men in the oldest Semitic
communities, takes the clan as the earliest social unit.^
For historical times this view is amply justified by the

1 The Religion of the Semites, 2d ed., London, 1894, p. 35.
30

PRIMITIVE SEMITIC SOCIAL LIFE 31

evidence, and there is much reason to believe, as we shall
see, that it extended far back into prehistoric times. In
Babylon it is true that at the very dawn of history cities
had superseded, at least in form, the communal clan
organization; but this was due to the fact that the rich
soil and abundant water of the Tigris-Euphrates valley
made agriculture easy, and paved the way for a civilization
which gradually outgrew the tribal stage. Whether the
Semites originated this civilization or borrowed it, has no
bearing on the fact that they were influenced by it. In
the earliest times, however, the town life did not materi-
ally modify the communal clan life. Cities, whether in
Babylonia^ or in Palestine,^ were at first simply fortified
dwellings of clansmen.

Through many sources clan organization may be traced
in the other parts of the Semitic domain. Thus, in the
genealogical lists of the Old Testament, we can trace, as
scholars now generally recognize, the clans of which the
Israelitish tribes were composed, since the writers, in
accordance with the patriarchal ideas of their own times,
have personified the nation as a man, tribes as his sons,
and clans as his grandsons or descendants. Sometimes
these clans can, with probability, be traced in extra-
Biblical sources, as Heber and Malkiel, clans of the tribe
of Asher, in the El-Amarna tablets;^ but whether they
can be so traced or not, there is no doubt but that the
names represent clans. Clans not mentioned in the Bible,
which once dwelt in Syria and Palestine, are also men-
tioned in the El-Amarna correspondence, such as the " Sons

I Of. Winckler's Altorientalische Forschungen, 1st ser., p. 232 ff.

* See the remarks of Robertson Smith in the Journal of Philology,
Vol. IX, p. 92.

' They appear in several letters as Mahiri and Malki-ilu. For the
Hehrew names cf . Gen. 46'', Nu. 26^^^ j Chron. V^i. For the cuneiform
references, cf. Schrader's KB., Vol. V, pp. 302-313. On the identification,
cf. Jastrow, in JBL., Vol. XI, pp. 119-122. Reisner {JBL., Vol. XVI,
pp. 143-145), following HaI6yy, makes them Cassites, but this does not
seem to me so probable.

32 SEMITIC ORIGINS

of Ebed-Ashera,"i " Sons of Labapa,"^ " Sons of Arzawa,"'
etc. These letters give us a picture of a seething mass of
clans, each struggling for the supremacy of Palestine.

Such lists as that of Gen. 36 attest the same organ-
ization for ancient Edom. The clan organization of the
whole of Arabia is attested in many ways. One evidence
of it is the list of names in Gen. 25^^^; while proof,
if possible of a still more convincing character, is found
in the Aramaic inscriptions brought from Hegra and
vicinity.* Claudius Ptolemy, in his "Geography," in
describing Arabia, gives a long list of nations which
must, as Robertson Smith pointed out, have been clans in
an easy state of flux, for by the time of Mohammed their
names had all disappeared.^ The existence of clans for
this region is also attested by the Sabsean inscriptions,
which have in recent years been recovered.^

Wherever in the dawn of history we can catch glimpses
of the Semites before the life of the cities had obliterated
more primitive traits, the clan is the unit of organization.

In Arabia, where, to the present time, the physical con-
ditions make a high degree of civilization impossible for
a large portion of the inhabitants, the same social organiza-
tion prevails. While the tribe has become the larger
unit of organization in the community, the conditions of
existence are such that clans, often exceedingly small,
live, move, and act together under the guidance of their
own sheik, who holds a position of somewhat ill-defined
subordination to the sheik of the tribe. '^ This organiza-
tion is so ingrained into the constitution of Arabic life,

1 KB., Vol. V, pp. 154, 155, e.g. Perhaps afterward the tribe of
Asher. See Chapter VI.

2 Ibid., pp. 306, 307, and 310-311.
8 Ibid., pp. 310-311.

< See CIS., Pt. II, Vol. I, Nos. 197, 198, 209, 215, and 221.
* See his Kinship, eta., p. 2.39.

' Cf. CIS., Pt. IV, Vol. I, Nos. 2", 292'8 (emended text), 406, 41*, etc.
' Cf. Doughty's Arabia Beserta, Vol. I, pp. 136, 251, and chs. ix-xii
and xiv-xvi, passim.

PRIMITIVE SEMITIC SOCIAL LIEE 33

that it has suryived the influences of Islam, of migration,
and of exposure to civilization, and still appears wherever
Arabs are found to-day. Clan organization exists, or has
existed in all parts of the world, and is recognized by-
sociologists as the simplest and earliest form of social
integration. 1 We may assume, therefore, that it existed
among the primitive Semites in their united home.
Indeed, if Keasbey is right in his estimate of the forces
which have made man social,^ Arabia never could have
supported a population of any size without it. A semi-
agricultural cultivation of the palm in the oases was, as
will be shown below, the chief food supply of the Arabs
almost, if not quite, from the time of their settlement in
the peninsula. No company of men could gain possession
of an oasis and hold it for cultivation without organiza-
tion for defence. Such an oasis would not support them
the year around ; they must either hunt or keep flocks and
herds. In Arabia, as will be pointed out below, there was
little hunting. If flocks and herds were kept, they must
be led forth to pasture. While some were cultivating the
oasis, others must take the more dangerous part of leading
the flocks and herds out to graze. This latter task
would naturally fall on the younger and more hardy men.
Keasbey has shown that it was under such conditions that
clans, or artificial brotherhoods, were formed. ^ We have
present, therefore, in Arabia from the start those economic
conditions in which the clan is forced into existence, and
may rest assured that we are treading on firm ground when
we assert that the clan organization was a part of primitive
Semitic life.

Giddings * regards the clan as having for its nucleus an
actual group of brothers and sisters, who form a totemic

1 See, e.g. Giddings's Principles of Sociology, p. 258 ff.

2 See his article, "The Institution of Society" in the International
Monthly, Vol. I, pp. 355-398.

« Ibid., pp. 385-398.

* Principles of Sociology, pp. 270-272.

34 SEMITIC OEIGINS

kindred and constitute a household; Keasbey,^ as a brother-
hood, artificial in organization, though not necessarily of
different stocks, who have selected the totem as a kind of
Shibboleth. Whatever the beginning of such a brother-
hood, its development is admirably sketched by Giddings.^
It forms at first an economic group, vi^ho aid one another
in obtaining food and redressing wrongs. The kinship of
such families is usually, among savages, reckoned through
the mother and not through the father. At a time too
remote for us to detect the origin of the practice, says
Giddings, natural brotherhoods are, by expulsion and
adoption, arising doubtless from economic causes, con-
verted into artificial fraternities; according to Keasbey
they were such from the first. These brotherhoods ac-
quire, in the animistic stage of culture, a peculiar sanc-
tity through the belief that men are akin to supernatural
beings. The belief that the individual is akin to his
totem reacts on his conception of human relationship;
and in time, though the members of a family may have
individual totems, the household regards itself as a unit,
and comes to have its collective totem in addition to these.
Adoption, then, becomes a more sacred ceremony ; exogamy,
if practised before for other reasons, now becomes a reli-
gious obligation, since it is sanctified by the totem ; thus,
all the practices of the fraternal group assume a more ob-
ligatory character. From time to time, the members of
such a household would encounter others who had acci-
dentally hit upon the same totem. These, they reason,
must be their brothers and sisters, since they are kindred
to the same totem as themselves. Thus, the brotherhood,
with all its privileges, rules of marriage, and obligations
for mutual protection, is enlarged. In a generation or
two there exists in such a group all varying degrees of
kinship, and the totemic clan is complete.

These descriptions of the genesis of the clan in general
may be taken as a tolerably accurate description of the
1 Op. cit., p. 393 ff. 2 Op. cit., pp. 270-272.

PRIMITIVE SEMITIC SOCIAL LITE 3a

Semitic clan. The proof of this on the social and eco-
nomic side will appear as we proceed. We may note now,
however, some of the proofs of Semitic totemism. These
have been more fully presented by the late Eobertson
Smith than by any other writer.^ He regards the proof of
totemism complete when we find : (1) " Stocks named after
plants and animals ; (2) the prevalence of the conception
that the members of the stock are of the blood of the
eponym animal, or are sprung from a plant of the species
chosen as totem; (3) the ascription to the totem of a
sacred character, which may result in its being regarded
as the god of a stock, but at any rate makes it to be re-
garded with veneration, so that, for example, a totem
animal is not used for ordinary food."^ Taking these as
guides. Smith found many Arabic tribes bearing the names
of animals as stock names, ^ and many traces in the Old
Testament of the same thing.* These names form a strik-
ing and impressive list, and form, with the names derived
from gods, a large proportion of Arabic personal and stock
names.® He also found that many Arabs believed them-
selves to be descended from such animals as the fox, wolf,
and hyena, and that some of them bewailed a dead gazelle
as a relative.® The third link in the proof is nearly want-
ing because of the veil which Mohammedan sources draw
as far as they can over the old heathenism. The nearest
approach to it which he found in Arabia was the existence
of two or three gods in animal form. Thus, Yaguth, the
lion-god, was worshipped in the time of the prophet by

1 In an article in the Journal of Philology, Vol. IX, entitled " Animal
Worship and Animal Tribes among the Arabs and in the Old Testament,"
and in his Kinship and. Marriage in Early Arabia, ch. vii. He has
been followed by Jacobs in his Studies in Biblical Archeology, London,
1894.

2 Kinship, p. 188.

» Journal of Philology, Vol. IX, pp. 79-88, and Kinship, pp. 192-201.
« Journal of Philology, Vol. IX, pp. 89-100.
6 Of. Smith's Kinship, p. 202.
• Ibid., pp. 203-205.

36 SEMITIC ORIGINS

several different tribes. Ya'uq, according to commen-
tators, was an idol in the form of a horse, while Nasr was
said to have the figure of a vulture.^ This is as near as we
can come to direct proof. As Smith pointed out, there is
some indirect proof in the fact that the jinn, who are said
in the Qur'an (Bi"") to have been partners with God, and
therefore probably old deities degraded, are generally
conceived in monstrous and hairy forms. ^

Among the Hebrews the survivals of totemism do not
form so complete a chain of proof for its existence, but
there are a number of sporadic traces of it in the Old
Testament which confirm the argument drawn from the
Arabic material. There are a number of animal names,
like Leah, Rachel, and Caleb, which were borne by reputed
ancestors of clans ; and in Ezekiel (8^") and Isaiah (66")
we have traces of an old animal worship revived in times
of distress, which seems to be a survival of totemistic
deities.^

The Old Testament affords some evidence that similar
conceptions were entertained by neighboring Semitic
tribes. Thus Oreb and Zeeb (the Raven and the Wolf)
are, in Judges (7^^), the names of Midianitish chieftains.
So Epher (the fawn or calf of the wild cow) is a Midian-
ite, Judsean, and Manassite clan.* Many others might be
cited if space permitted the reproduction of the investiga-
tions on this subject.^ Although the proof from the Old
Testament is not so complete as from Arabia, yet two out
of the three necessary classes of evidence are found.

1 Cf. Kinship, pp. 208, 209, and Wellhausen's JReste arabische Heiden-
tums, 2d ed., Berlin, 1897, pp. 19-23.

2 Kinship, p. 211.

' Cf. Smith's Beligion of the Semites, 2d ed., p. 357, Cheyne's " Isaiah,"
in Haupt's SBOT., p. 200, n. 5, and Toy's "Ezekiel," in SBOT., p. 110, n. 7.

* Cf. Gen. 25*, 1 Chron. 4", 6^, and Robertson Smith in the Journal
of Philology, Vol. IX, p. 91. See also Gesenius, Handworterbuch, 13th ed.,
Leipzig, 1899, which follows Smith.

6 Cf., in addition to Smith and Wellhausen, the list in Jacohs's St'udies
in Biblical Archoeology.

PRIMITIVE SEMITIC SOCIAL LIFE 37

In other Semitic countries occasional sporadic traces
are found which point in the same direction. Thus,
among the Guti or Suti, a tribe on the east of Babylonia,
the goddess Ishtar was represented in the form of a lion
or riding on a lion.^ In the Gilgamish epic, which, as
we now have it, centres in the city of Erech in southern
Babylonia, Ishtar marries now a bird, now a lion, now a
horse, and tries to marry a man,^ — facts which point to a
totemistic circle of ideas. At Eryx, in Sicily, Ashtart
was thought to have the form of a dove,^ and at Tyre, the
head of a bull.* It is to such a conception that the book
of Tobit (1^) alludes when it tells how people sacrificed
to "She-Baal, the cow."^ In the same class of evidence
we should probably put the calves at Bethel and Dan
which were said to be images of Yahwe (1 Kgs.

1228).

These reasons, slight as they may seem, come from
widely different parts of the Semitic territory, and are, for
that reason, significant. Sporadic traces of totemism so
widely scattered can only be explained by supposing that
the primitive clans in their old Arabian home were totem-
istic. Indeed, it is possible that this stage had been
reached before the Hamites separated from them, for the
well-known animal worship of the Egyptian nomes, in
which each nome worshipped a different animal, is posi-
tive proof of the existence of totemism among that people.®

1 Cf. Hehraiea, Vol. X, pp. 26, 27. This and other evidence of a simi-
lar character was collected in my " Semitic Ishtar Cult," puWished in
Vols. IX and X of Hebraica.

2 Cf. Hehraiea, Vol. X, p. 5, and Jastrow's Beligion of Babylonia and
Assyria, Boston, 1898, p. 482.

2 Cf. jElian, Be Natura Anamalium, IV, also Hebraica, Vol. X, p. 49.

* Cf. the extracts from Philo of Biblos, published by Orelli as Sanchn-
niathonis Fragmenta, p. 30, really taken from the Prmp. Evang. of
Eusebius, Bk. I, 10, 31, Josephus's Antiquities, 8, 5, 3, Against Apion,
1, 18, also Hebraica, Vol, X, p. 31.

^ Tj BtiaX Trj da/idXei.

8 The attempt of Robert Brown, Jun., Semitic Influence in Hellenie
Mythology, Williams & Norgate, London, 1898, p. 56 ff., to prove that

SEMITIC ORIGINS

The early culture of Egypt embalmed this system for our
study as the amber of the Baltic sometimes embalms a fly;
and the fact that the totemic system of thought appears
both among the Semites and the Hamites makes it pos-
sible that before the separation of the two branches of the
Hamito-Semitic race the totemic clan had already been
developed. This possibility will again confront us at a
later point; but however it may be determined, we are
justified in holding that the totemic clan was the primitive
Semitic social organization.

The economic purpose for which the clan organization
was formed by the primitive Semitic folk was the defence
of their date-growing oases and their domestic animals
in their pasture lands, or for the attack of similar pos-
sessions of their neighbors. In a country where the con-
ditions of life are as hard as they are in Arabia, the
population has again and again, far back into prehistoric
times, become too numerous to be supported by the sterile
soil. Some who were pushed away from the oases or
better pasture lands were compelled to plunder others for
a living, until at last the pressure from within forced a
wave of emigration through some convenient channel into
another territory. These conditions down to the present
time force many Arabs to become robbers, and make bands
armed for plunder the terror and often the destruction of

totemism never existed in Egypt, (1) because no real Egyptologist believes
it, and (2) because Strabo (XVII, 40) says that all worshipped the ox,
cat, hawk, and ibis, is a signal failure. Were it true that no Egyptologist
believes totemism to have existed in Egypt, it would only prove that
Egyptologists allow themselves a narrow range of studies, so that the
facts of anthropology escape them. It is doubtful whether this charge
can justly be brought against them. Cf. e.g. Maspero's Mtudes de
Mythologie d'Archeologie Egyptienne, Paris, 1893, Vol. II, p. 277.
Maspero's statement that their myths are of the same sort as those
of the savages of the old and new world, is very like a confession
of a belief in totemism. Strabo was a late writer, and, by his time,
political unity had created syncretism so that the gods of some locali-
ties had obtained universal recognition ; accordingly his statement
proves nothing.

PRIMITIVE SEMITIC SOCIAL LIFE 39

the nomad's life,^ and must have operated among the
primitive Semites much as they do to-day. Thus, then
as now, clans must exist for mutual protection. Some of
these would settle on an oasis, and their older and weaker
men would aid the women in cultivating the date-palm,
while the more hardy of the men led the small flocks and
herds out into the neighboring pasture lands.^ Those
who were so unfortunate as to obtain no oasis would wan-
der up and down with their flocks and herds seeking
pasturage and plunder as opportunity offered.

In theory, such economic conditions should produce all
three classes of clans : the communal, for the cultivation
of the oases ; the republican, for defence and for caravan
trade; and the pastoral, for the part of the population
which could obtain no oasis. All these types were, as we
shall see, in time produced, but the dependence of all
classes on the oases would, in Arabia, long hold the for-
mation of the republican and patriarchal type of clan in
check, and enable the communistic clan to exert such an
influence as to leave its stamp upon the organization of
society. Indeed, it is probable that for unnumbered cen-
turies the clans which were deprived of the privileges of
oases found life so hard that the conquest of other coun-
tries became for them a necessity, even before the repub-
lican or patriarchal type had become fully fixed. These
facts will come out more fully as we proceed.

When we go a step farther back and seek to determine
the constitution of the primitive Semitic family, we are
met by greater difficulties. Those who have labored in
this field hitherto have worked with the sociological
theory of McLennan,^ who held that in the primitive con-

1 Of. Doughty's Arabia Deserta, Vol. I, pp. 345, 489, 505, etc. ; also
Sale's The Koran, p. 24.

!" Of. Payne's History of the New World called America, Vol. II,
p. 7 fE., and Keasbey, in the International Monthly, Vol. I, p. 390 ff., for
analogous examples.

' Cf. G. A. "Wilken's Het Matriarchaat bij de Oude AraUeren, Am-
sterdam, 1884 ; W. R. Smith's Kinship and Marriage in Early Arabia,

40 SEMITIC OEIGmS

dition of man the relation of the sexes to one another was
one of unrestrained promiscuous intercourse; that this
was succeeded by a state of polyandry, and this, in turn,
by the practice of polygamy, out of which monogamic
marriage has grown. ^ This view presupposed that in the
development of society the relation between the sexes had
everywhere advanced according to one general law. In
the polyandrous state of society it was found that kinship
was reckoned through the mother, and it was inferred that
woman, and not man, was the head of the clan. Thus, it
was supposed that a matriarchate everywhere preceded
a patriarchate, and that in the evolution of society the
relative position of the sexes has been reversed.

More recent investigators of social problems are, how-
ever, unanimous in the opinion that polyandry is not a
social condition through which all mankind has passed,
but a phenomenon of social evolution which, under very
special conditions, has appeared among a few races only.^
The evidence on the matter seems overwhelmingly in
favor of the latter view. It cannot, however, be repro-
duced here, but must, to be appreciated, be studied in the
works of special students of the subject.^ It appears,
therefore, that the real matriarchate is comparatively rare.
Among a few peoples, like the Nairs of the Malabar coast, it
seems really to exist,* and in such families as those of the

Cambridge, 1885 ; and an article of the writer's, " The Kinship of Gods
and Men among the Early Semites," in the JBL., Vol. XV, p.
168 ff.

^ See, e.g., MoLennan's Studies in Ancient History, Macmillan & Co.,
1886, oh. viii.

2 See Spencer's Principles of Sociology, Vol. I, p. 660 (Am. ed., New
York, 1897) ; Staroke's Primitive Family, New York, 1889, p. 139 ff. ;
Letourneau's The Evohition of Marriage and the Family, New York,
p. 320 ff. ; Westermarck's The History of Human Marriage, Mac-
millan & Co., 1891, pp. 459 and 505-508; Lubbock's The Origin of
Civilization and the Primitive Condition of Man, 5th ed., New York,
1892, p. 143 ff.; and Giddings's Principles of Sociology, pp. 155 and 276.

8 Ibid.

* Cf. Reclus's Primitive Folk, New York, 1891, p. 165.

PRIMITIVE SEMITIC SOCIAL LIFE 41

Andaman islanders ^ it would be the natural outcome of
the unorganized condition of society. It frequently hap-
pens where polyandry is practised that the brother of the
mother is the head of the family, and rears his sister's
children, so that there is an avunculate rather than a
matriarchate.

The late W. Robertson Smith, the great investigator of
this phase of Semitic life, in whose tracks others of us
have followed at a distance, and "non passibus sequis,"
based his investigations on the theory of McLennan ; and
though many of his results are permanent, and the mass
of material he collected invaluable, yet, in the light of the
present state of the science of sociology, the whole subject
merits a new examination.

Westermarck ^ and Giddings^ appear to be right in
holding that the family of primitive man was an inter-
mediate development between that of the highest animals
and the lowest living men. In the lowest existing human
societies the usual form of marriage is a temporary mo-
nogamy.* It is improbable that back of this there was a
time when the marriage relations, taking mankind as a
whole, were less clearly defined, since temporary mar-
riages of this character appear among the higher apes.^
It is possible — and a possibility of which writers on soci-
ology take too little notice — that increased intelligence
on the part of man may in some races have introduced into
sexual relations degenerate practices. Higher mental
power is, in the first instance, usually devoted to increased
gratification of appetite, until the growth of moral senti-
ment brings the power of intelligence under the sway of
worthy aims. It may easily have been, therefore, that
human intelligence first, in some instances, exercised

1 See Giddings's Principles of Sociology, p. 266.

2 History of Human Marriage, pp. 14, 15, and 50.
8 Principles of Sociology, p. 264.

4 Ibid.

' History of Human Marriage, pp. 14, 15, and 50.

42 SEMITIC ORIGINS

itself in gaining more frequent gratification for sexual
desire. It is not likely that this would overthrow the
kind of family organization which had obtained among
our prehuman ancestors, since the economic necessities
of life would be sufficient, in most parts of the world, to
insure the union of father and mother until the mother
and child could obtain food for themselves. It would,
however, tend to produce lawlessness in sexual relations,
which would bring into existence a sort of promiscuity by
the side of the primitive temporary monogeimy. This
promiscuous intercourse would, on the part of the men, be
participated in by those of all ages; while among the
women it would be more often the young who would
indulge, for these would more often attract by their beauty,
— which in such communities quickly fades, — and both
desire and inexperience of the consequences would lead
them in this direction. This freedom has, I think, been
perpetuated among those peoples who attach a religious
significance to an act of free love on the part of their
women before marriage.

That something like this occurred in the develop-
ment of the Semites, two facts make probable: (1) the
tendency of the early Semitic peoples to sexual excesses,
a trait which points to an early bent in this direction ; and
(2) the fact that in the Ishtar cult in several different
Semitic countries a religious importance was attached to an
act of free love on the part of woman before she entered wed-
lock. Thus, Herodotus, Strabo, and the apocryphal letter
of Jeremiah tell us that every Babylonian woman must once
in her life offer herself in the temple of the goddess to
whatever man might come.^ Ephraim,^ the Syrian, affirms
the same practice among the Arabians in their worship of
the mother goddess; and Augustine, in describing the

^ See Herodotus, Bk. I, oh. 199 ; Strabo, Bk. XVI, ch. l^o, and the
Epistle of Jeremiah, vv. 42, 43. Cf. Hebraiea, "Vol. X, pp. 20, 21, and
JBL., Vol. X, p. 79 ff.

2 Opera, Vol. II, pp. 458, 459. Cf. Hebraiea, Vol. X, pp. 58, 59.

PRIMITIVE SEMITIC SOCIAL LIFE 43

feast of the mother of the gods among the Carthaginians,
makes it probable that the custom also existed there. ^
Herodotus, in the passage cited, says that the custom is
also found in parts of Cyprus. We know that a kindred
goddess was worshipped there,^ and it was no doubt in
connection with her worship. Lucian vouches for the
existence of the custom at Byblos, the old Phoenician
Gebal, but tells us enough to show that it had there under-
gone certain modifications ; a woman who did not wish to
sacrifice her chastity might sacrifice her hair.^

This custom, thus widely extended, is pretty good proof
that the practice in question goes back to primitive Semitic
times. This view is confirmed by a passage in the Gil-
gamish epic. In the second tablet of this poem there is a
description of how Eabani, a wild man of the mountains
or a primitive man, was enticed from the beast with which
he had previously satisfied his passion,* by an emissary of
the goddess Ishtar.^ The fact that primitive man is here
regarded as having promiscuous intercourse with the ani-
mals is in itself a testimony to its existence among human
beings. Jastrow holds that there is a reflection of this,
or a similar story, in Gen. 2, where he believes the
original form of the narrative represented man as having
intercourse with the beasts until woman was brought to
him; and that then he abandoned the animals and became
"one flesh " with her.^ This view has much to commend

1 See his Be Civitate Dei, Bk. 11, ch. iv. Cf. Hehraica, Vol. X, pp. 50, 51.

2 Cf. Hebraica, Vol. X, pp. 42-47, and Journal of Hellenic Studies
for 1888, pp. 175-206.

" Lucian's Be Syria Bea, § 6 ff.

* That this is the meaning of the passage Jastrow has pointed out,
AJSL., Vol. XV, p. 202.

» See Haupt's Nimrodepos, Leipzig, 1884, pp. 10, 11 ; Jeremias's Izdu-
bar-Mmrod, Leipzig, 1891, pp. 15-18 ; Hehraica, Vol. X, pp. 2, 3 ;
Jastrow's Religion of Babylonia and Assyria, pp. 477, 478 ; and Jensen,
In Sohrader's KB., Vol. VI, Berlin, 1900, pp. 125-127. According to
Jensen's reconstruction of the poem, the passage is in the first tablet.

" See Jastrow's article, "Adam and Eve in Babylonian Literature,"
AJSL., Vol. XV, pp. 207, 208.

44 SEMITIC ORIGINS

it. The presence in the Gilgamish epic of a female
priestess whose life is consecrated to this impure service
reveals the existence of an institution which could hardly
fail to grow out of the conditions which we have supposed
to exist. From the earliest times there must have been,
as there are now, women who, for one reason or another,
gave their lives to the satisfaction of desire. Under
primitive conditions this would be done much more freely
than now, since there could not be much, if any, public
opinion against it. Such religion as these early men had
would, in the lapse of time, preserve, through the con-
servatism of mankind with regard to religious practices,
these conditions under the guise of sacred service far
beyond the state of society in which they had their birth ;
and thus present the anomaly which we find in so much
of ancient Semitic life, of an impure priestess ministering
in a community whose marriage ideal was relatively pure.
The Eabani episode is one of the oldest strata ^ of a
poem, the later parts of which are some four thousand
years old, and may well be held to reflect tolerably primi-
tive ideas. We cannot be far wrong, therefore, if we
hold, on the evidence presented, that in one of the earliest
stages of Semitic development — a stage reached perhaps
before their separation from the Hamites — such a strong
tendency to unregulated intercourse existed, and that
its results are seen in the religious practices which sur-
vived here and there far down into historic times. In
Oman, where Mohammedan influences are felt less than
in the most of Arabia, maiden virtue is, according to Pal-
grave, still of little account.^ Sprenger cites a curious
passage from Yaqut,^ with regard to the town of Mirb§,t,

1 Cf. Jastrow's Religion of babylonia and Assyria, pp. 474-478, 513.

=! Cf. Palgrave's Central and Eastern Arabia, Vol. II, p. 267.

' Cf. Sprenger's Alte Geographie Arahiens, p. 97. For the original
see Jacut's Geographisches Worterbuch, ed. "Wilstenfeld, Leipzig, 1869,
Vol. IV, p. 482. Noldeke also admits {ZDMG., Vol. XLV, p. 1.55),
that among the Semites a kind of prostitution was practised without
shame.

PRIMITIVE SEMITIC SOCIAL LIFE 45

in the course of which it is said : " Their women go each
night to the outer part of the city and devote themselves
to strange men, and sport with them the greater part of
the night. The husband, brother, son, and nephew goes
by without taking notice, and entertains himself with
another." This must have been a survival, under some-
what changed conditions, of the primitive tendency of the
Semites to unregulated indulgence.

That there existed a temporary monogamy, such as
sociologists postulate for the earliest human families, side
by side with this unregulated intercourse, can also be
shown to be true. Whether living in an oasis or wander-
ing from place to place in the deserts of Arabia, women
would be, from the earliest times, needed to perform the
drudgery of the household and the camp, that the men
might be free for those duties everywhere considered more
manly by savages and barbarians — the duties of fighting
for defence or plunder. These women must have been, in
the earliest period, the mothers and sisters of the men, and
not their wives, for ancient Semitic marriage was every-
where exceedingly temporary and divorce extremely com-
mon, — facts which show that the primitive Semitic
marriage tie was an evanescent bond. These facts are
abundantly attested by the Old Testament, the Baby-
lonian contracts, the Qur'an, by numerous instances in
Arabic life, and by the condition of Abyssinian society at
the present time.

Among the Israelites of the Old Testament the senti-
ment seems to have been somewhat against divorce ; and
yet the law of Deuteronomy ^ makes it so exceedingly easy
that it evidently points back to a time when divorce was
much more common.

Among the Babylonians the frequency of divorce is not

so easy to trace, since we have not, as in Deuteronomy,

general statements of law, but must draw our inferences

from the study of special cases. Nevertheless, in the few

I Deut. 241-3. cf. Isa. 50i.

46 SEMITIC OKIGINS

marriage contracts and records of Babylonian divorce
which have been studied, a sufficient number of instances
appear to make it clear that divorce was not uncommon.
Peiser has pointed out that two tablets in the British
Museum reveal, upon comparison, that a woman who had
been married to one man was within eight months married
to another, while the first was still living. ^ The fact,
too, that provisions for divorce were usually introduced
into the marriage contracts of those women who married
without a dowry, is clear proof that divorce was so com-
mon in Babylonia that women were compelled to protect
themselves against it in the marriage contract.^ Where
the woman carried to the husband a dower, this was not
necessary, since in Babylonian law the dowry was always
hers, so that in case the husband divorced her he would
lose it. In such cases the self-interest of the husband was
thought to be a sufficient protection to the wife.^

The evidence from the Arabs is more abundant and,
from sources both ancient and modern, is of the same
character. The Qur'an contains two passages which attest
the frequency of divorce. Sura 65i"^ takes it for granted
that divorces will be frequent, and provides that the
woman shall not be sent forth burdened with the prospects
of motherhood; while Sura 33*^ supposes that men may
frequently divorce their wives for whims after marrying
them, but before marriage relations have really been estab-
lished. The custom of divorce for any cause, at the
wish of the husband, was, in the time of the prophet, too
thoroughly fixed in Arabic custom, and too congenial

1 See his Babylonische Mechtsleben, Berlin, 1890-8, Vol. II, pp. 13-15.
The first of these texts is published in Strassmaier's Babylonische Texte,
Heft VII, No. Ill ; the second is, so far as I know, unpublished.

2 Cf. Strassmaier, ibid., No. 183, and Inschriften Nabuchodonosor, No.
101 ; also, Peiser, op. cit., Vol. IV, pp. 12, 13 ; and Merx in Beitrage zur
Assyriologie, Vol. IV, pp. 4-8 ; and my article on "Contracts," §§ viii,
ix, in Assyrian and Babylonian Literature, Aldine ed., New York, 1901.

8 See my remarks in the article on "Contracts," § ix, in Assyrian
and Babylonian Literature.

PRIMITIVE SEMITIC SOCIAL LIFE 47

to the natures of the prophet and his followers, to be
changed; hence it was crystallized in Mohammed's law
and passed on to other generations. ^ This liberty has
been fully exercised by many of the faithful. Thus Ali,
the son-in-law of the prophet, married, including all that
he married and divorced, more than two hundred women.
Sometimes he included as many as four wives in one con-
tract, and divorced four at one time, taking four others in
their stead. ^ A certain Mughayrah b. Sha'abah is reported
to have married eighty women in the course of his life,^
while Mohammed b. At-Tayib, the dyer of Baghdad, who
died in the year A.H. 423, at the age of eighty-five, is
said to have married in all more than nine hundred
women. If he began his marital career at the age of
fifteen, he must have had on the average nearly thirteen
new wives a year through his whole life.* This liberty
is exercised in Arabian countries still. Palgrave relates
that the Sultan of Qatar in eastern Arabia married a new
wife every month or fortnight, on whom the brief honors
of matrimony were bestowed for a like period, and who
was then retired on a pension.^ Doughty also tells® how
Zeyd, his host, a petty sheik of the Bedawi, not only
permitted one of his wives to be courted by another Arab,
but offered to divorce her that Doughty might marry her.
Indeed, in parts of Arabia divorces are, in certain cases,
not necessary, since the marriages are contracted for a
limited period of definite length. Ammianus Marcellinus
(XIV. 4) gives this as their usual type of marriage. After
a certain day, he says, the wife may withdraw if she

1 For an excellent account of divorce among the Arabs, see Wellhausen
in the Nachrichten der Kgl. Oesell. d. Wiss. zu Gott., 1893, p. 452 ff.

^ Cf. Lane's translation of the Thousand and One Nights, Vol. I, p.
318 ff., cited by Wilken, Set Matriarchaat bij de Oude Arabieren, p. 18.

8 Ibid.

* Ibid.

s Central and Eastern Arabia, Vol. II, pp. 232, 233.

8 Arabia Deserta, Vol. I, pp. 320, 321. Zeyd had once before found
a husband for a divorced wife of his, see ibid. , p. 237.

48 SEMITIC ORIGINS

pleases. Somewliat of the same character is a temporary
form of marriage which still exists in Sunan, a town fifteen
days from Mocha in south Arabia. It is described as fol-
lows : " In all the streets there are brokers for wives, so
that a stranger, who has not the conveniency of a house in
the city to lodge in, may marry and be made a free burgher
for a small sum. When the man sees his spouse and likes
her, they agree on the price and term of weeks, months, or
years, and then appear before the Kadi (qSdhi), or judge
of the place, and enter their names and terms in his book,
which costs a shilling or thereabout. And joining hands
before him the marriage is valid, for better or for worse,
till the expiration of the term agreed upon. And if they
have a mind to part or renew the contract, they are at lib-
erty to choose for themselves what they judge most proper;
but if either wants to separate during the term limited,
there must be a commutation of money paid by the separat-
ing party to the other according as they can agree ; and so
they become free to make a new marriage elsewhere." ^

In Mecca, whither throngs of pilgrims regularly resort,
some of whom tarry for longer or shorter spaces of time,
marriages of similarly short duration are still entered
into; and women go thither from Egypt with the avowed
purpose of entering into such alliances.^

In Abyssinia civil marriages, into which ordinary people
enter, are still dissoluble at will, and divorce is very fre-
quent. It is nothing unusual for husbands and wives to
exchange partners, all remaining as before on the best of
terms. Sometimes marriages are contracted for a fixed

1 Quoted by Wilken in Het Matriarchaat bij de Oude Arabieren, p. 15,
from Hamilton's N'ew Account of the East Indies, Vol. I, pp. 52, 53.

2 See C. Snouok Hurgronje's Mekka, Haag, 1888-9, Vol. II, p. 5 ff.,
and 109-112, and S. M. Zwemer's AroMa, the Cradle of Islam, New
York (Revell), 1900, p. 41. Zwemer is, however, dependent on Hur-
gronje. In Somaliland, where the native customs have been shaped by
Arabic immigration, till it is not easy to tell always how much is native
and how much is not, divorce is very common. Cf. Siidarabische Expe-
dition, Bd. I, Die Somali- Sprache von Leo Reinisch, Wien, 1900, p. 109.

PRIMITIVE SEMITIC SOCIAL LIFE 49

period, at the end of which husband and wife separate.
No stigma attaches to those who find a change of partners
desirable. Inconstancy is common and chastity not highly
valued. 1

When now we find in all Semitic countries a tendency
to make the term of marriage brief, — a tendency which it
requires a high degree of civilization to subdue in them,
— the inference is surely valid, that among the primitive
Semites marriage relations were in like degree temporary.
It is contrary to all analogy to suppose that the affections
of the primitive Semite were more constant than those of
his semicivilized descendant, or that there were in ancient
times stronger inducements than in more recent centuries
for the perpetuation of the marriage tie.

This fact is one of great importance, since its effect
upon the constitution of the primitive Semitic family
must have been serious. When marriages were of brief
duration, and the same man had several wives in succes-
sion, the most of them cannot have been his sisters, even
if such marriages had been permitted. There are, as we
shall see by and by, some possible instances of such mar-
riages among the Semites ; but for the most part the feel-
ing against mating with members of the same family,
which is so widely disseminated among the races of the
world,^ appears also among the Semitic peoples. Even if
this feeling had been absent, the transitory character of
marriage and the frequency with which men took new
wives would make it certain that most of them would be
of other families, if not of other clans.

These wives would, when discarded, return to their
kindred, if indeed they had ever left it, and would there,

1 Cf. Bent's Sacred City of the. Ethiopians, 1893, pp. 31 and 35 ff., and
A Visit to Abyssinia, by W. Winstanley, London, 1881, Vol. II, pp. 73, 74,
also Modern Abyssinia, by Augustus B. Wylde, London, 1901, pp. 161
and 254.

2 See Starcke, Primitive Family, pp. 210, 211 ; Westermarck, History
of Human Marriage, pp. 544, 545; and Giddings, Principles of Sociology,
p. 267.

B

50 SEMITIC ORIGINS

if they did not marry again, find support. Since the period
of a woman's life during which she was desired in mar-
riage was much shorter than the corresponding period in
the life of a man, many of these discarded wives must, in
any event, have been ultimately left with their own kin-
dred ; 1 where, if they carried their children with them, they
would be esteemed, on account of the children, as the real
perpetuators of the clan. It is therefore altogether prob-
able, as was remarked above, that the women who, in the
primitive Semitic clan, performed the drudgery, whether
in oasis or in desert life, were usually the sisters and
mothers, and not the wives, of the men.

Before, however, we accept this conclusion, with all its
consequences, it is necessary to examine two other points
which are closely connected : (1) the residence of the wife
during her marriage; and (2) the method of reckoning
kinship. In marriages of a temporary nature four differ-
ent cases are possible: (1) the wife may live with her
husband's kindred while married and return to her own
when divorced, he retaining the children; (2) she may
live while married with her husband's kindred, but on
returning to her own take the children with her; (3) she
may live in her own clan, whither the husband goes to
live with her, she retaining the children when he with-
draws ; and (4) she may reside in her own clan and the
husband in his, simply receiving visits from him from
time to time, in which case the children remain with her.
In the first of these cases the children would belong to the
clan of the father, while in the last three they would
belong to the clan of the mother.

The first of these conditions is that which has prevailed
in the Arabian world from the time of the prophet to the
present. The Qur'an (Sura 65^) specifically provides
that the children shall be reared for the father, and at his
expense. Many interesting instances of this might be
cited in the history of Mohammedan families ; for exam-

1 Cf. Giddings, op. cit., p. 264 ft.

PRIMITIVE SEMITIC SOCIAL LIFE 51

pie, Zeyd es-Sheychan, the sheik who was Doughty 's host,
and to whom reference has already been made, divorced
the mother of his son Selim, but reared the son in his own
family.^ If this system of paternal kinship were primi-
tive, we might suppose that the Semitic family had always
existed in much the form of the Arabic family of to-day.
If, however, it can be shown that descent was once reck-
oned through the mother, and that the present patronymic
family has superseded a metronymic organization, we shall
then be at liberty to inquire which of the last three posi-
tions we supposed above to be possible actually represents
the status of the primitive Semitic wife.

The late Robertson Smith and others have established
the fact, as well as the state of the evidence will permit
it to be established, that back of the custom of tracing
descent through males there was a time when the Semites
traced it through females. ^ It is true that the first point
which Smith makes, that if kinship were reckoned by
blood it would have to be reckoned through the mother,
because in primitive times paternity was uncertain, —
owing to the state of promiscuity,^ — is one which, in the
light of recent sociological investigation * must be aban-
doned, for it is altogether likely that in most cases the
father was known. Of his arguments, which still remain
valid, a summary may be made as follows : (1) The well-
known Biblical phrase for relationship is "bone of my
bone and flesh of my flesh." "Flesh" is explained in
Lev. 25*8 by the word for clan. The Arabs attach great
importance to a bond created by eating together; we must
suppose, therefore, that the bond between those born of
the same womb and nurtured at the same breast would be

1 Doughty's Arabia Deserta, Vol. I, pp. 101, 217, 237, etc.

2 See his Kinship, pp. 145-165.
^Ibid., pp. 146-148.

* See Staroke's Primitive Family, p. 25, his article in the International
Journal of Ethics, Vol. Ill, p. 455, and Westermarck's History of
Human Marriage, pp. 108, 109.

52 SEMITIC ORIGINS

more nearly of the same "flesh " and the same "clan " than
any others. (2) The word rahim, womb, is the most gen-
eral word for kinship, and points to a primitive kinship
through the mother. (3) The custom called 'acica, by
which a child is consecrated to the god of his father's
tribe, cannot have been primitive. It must have sprung
up in a state of transition to insure the counting of the
offspring to the father's side of the house. (4) Cases
occur in the historical period in which a boy when grown
attaches himself to his mother's tribe. The poet Zohair
is a case in point, and Arabic antiquarians appear to have
known that such cases were not uncommon.^ (5) The
fear that sons would choose their mother's clans led men
who were wealthy to marry within their own kin.
(6) The relation between a man and his maternal uncle
is still considered closer than between a man and his
paternal uncle. (7) Joseph's sons born of his Egyptian
wife were not regarded as members of Israel's clan until
formally adopted by him (Gen. 48^-^). (8) Abraham
married his paternal sister, who was not the daughter of
his own mother. Tamar might have legally been the wife
of her half-brother Amnon, the relationship being on the
father's side (2 Sam. IS^^). Such unions were known in
Judah as late as the time of Ezekiel (see ch. 22"). Tab-
nith, king of Sidon, married his father's daughter,^ and
such marriages were known in Mecca. Since the marriage
of those really regarded as brothers and sisters was abhor-
rent to the Semites, kinship must in these cases have been
counted through the mother. (9) In the Arabic genea-
logical tables metronymic groups are still found. (10) In
Aramaic inscriptions found at Hegra metronymic clans
appear.^

Although these arguments of Smith are interwoven

1 Cf. Smith's Kinship, pp. 155, 246-253.

2 Cf. 018., Pt. I, Vol. I, No. 3, 11. 13-15.

3 Cf. CIS., Pt. II, Vol. I, Nos. 198 and 209. See also Smith's Kinship,
pp. 313-316.

PRIMITIVE SEMITIC SOCIAL LIFE 53

with some theories of polyandry, the consideration of
which must be postponed a little, and with some argu-
ments which do not appear to be valid, these which we
have summarized present facts which, regardless of any
theories of marriage, prove that at one time kinship was
reckoned through the mother.

This conclusion is corroborated by evidence gathered by
other scholars. Noldeke noted that in the religious texts
of the Mandaeans a man is described as the son of his
mother, which indicates that among them kinship was
reckoned through the mother. ^ Peiser has pointed out*
that among the Babylonians a man could if he chose re-
nounce his family and join the kindred of his wife, which
is a relic of the same custom. Wellhausen has observed '
that in the genealogies of the Pentateuch the J document
reckons descent through the mother, while in the P
document it is traced through the father.

These arguments may be confirmed by several important
considerations. If descent had not been reckoned through
the mother, the position which, as will be pointed out
below, woman held among the early Semites would have
been impossible, as would also a type of marriage for which
there is considerable proof, and which will be considered
in its place.

If, now, the fact be accepted that kinship was counted
through the female line, their habit in this respect is
found to conform to that of most other primitive peoples,*
and a vantage ground is obtained from which the social
phenomena which remain to be considered become intel-
ligible to us. If children did not belong to the clan of
the father, the first of the possible forms of marriage men-

1 Monatsschrift, 1884, p. 304.

2 MiUheilungen der vorderasiatische Gesellschaft, 1896, p. 155.

" Nachrichten d. Kgl. Oesell. d. Wiss. zu Gott., 1893, p. 478, n. 2.

* See Spencer's Principles of Sociology, Vol. I, pp. 698, 703 ; Lubbock's
Origin of Civilization, etc., 5th ed., pp. 151-157 ; Starcke's Primitive
Family, pp. 18, 25, 37, 39. 74 ; Westermarck's History of Human Mar-
riage, pp. 96, 97, 539 ; and Giddings's Principles of Sociology, p. 265.

54 SEMITIC ORIGINS

tioned above is clearly eliminated. The mother, as a rule,
when she left her husband's residence (if she had lived
there at all) must have taken the children with her; and
if she resided in her own clan, it is clear that she retained
the children. Of course it is possible that from the earli-
est times a man might, when the clans were well disposed
to one another, induce a woman to leave her own kindred
and go to live with his. Of this we have almost no evi-
dence, and in the nature of the case can obtain little.
The point proved by Smith, ^ however, that in early pre-
Mohammedan times the natural protectors of a woman
were not her husband and his kindred, but her own rela-
tives, makes it improbable that in the earliest Semitic
communities the woman left her own people at all. Prob-
ably, therefore, the second of our possible arrangements of
Semitic marriage should also be eliminated.

Of the third possibility — the residence of the husband
in the wife's, tribe — we have more direct evidence. The
classical instance of this, which all writers cite, is the case
of Jacob and his wives, Leah and Rachel.^ Jacob lived
with them in Laban's clan, and when he left was blamed
for taking away the children. Laban declared, "the
daughters are my daughters and the children are my
children" (Gen. 31*^), i.e. they belong to my clan. A
second argument, and one which proves that the case of
Jacob and Laban is not an isolated instance, is found
in the fact that the phrase for marriage which is used
throughout the Old Testament, which is found in Syriac,
and is still used in south Arabia, ^ is "he went in unto
her." Smith has shown* that this phrase originated when
it was the custom for a man to go to reside in his wife's
tent in her tribe. In Yemen it is still the custom for the
"going in "to take place in the bride's house; and the
bridegroom, if home-born, must stay some nights in
the bride's home, and if a foreigner, must settle with

J Kinship, pp. 101-103. ' Smith's Kinship, pp. 167, 168.

2 Gen., chs. 29-31. « Ibid., pp. 167-172.

PRIMITIVE SEMITIC SOCIAL LIFE 56

the tribe. ^ Smith also pointed out that the custom in
north Arabia which compels a man to build a new tent
for his wife, is an outgrowth of the older practice of enter-
ing the wife's tent. In the same region it sometimes
happens still that the wife refuses to leave her tribe, and
the husband is compelled to leave his and go and join
hers. Lady Blunt relates such an instance which came
under her own observation.^ That this also occurred in
ancient Babylonia, the case cited above of the man who
joined his wife's family is sufficient to prove. Such evi-
dence as this, coming from so many portions of the Semitic
territory, makes it clear that this kind of marriage was a
primitive Semitic practice.

In such marriages many circumstances might arise to
call the husband away and interrupt the marriage rela-
tions. The clans might become hostile, so that it would
be unsafe for him to remain, or his fancy might weary of
the bride's attractions, or of her people, and then he would
wander elsewhere to contract a similar alliance. Such
marriages are called beena marriage, the name given them
in Ceylon, where they were first studied. They are found
in many parts of the world. ^ The children in such cases
remain of course with the mother, are reared by her kin-
dred, and become a part of her clan. The net result,
therefore, of our discussion up to this point is the estab-
lishment of the fact that the primitive Semites practised
beena marriage, that the children belonged to the tribe of

1 Smith, ibid., p. 168. Among the Hamitic Somalia of East Africa,
among whom the Arabs have penetrated and by whom many Arabic
customs have been adopted (cf . Beitrage zur Mhnographie und Anthro-
pologie der SomSl, Galla und Harari, von Philipp Paulitschke, 2d ed.,
Leipzig, 1888, p. 2 f£.)i it is still customary, vrhen a young man marries,
for the bride, aided by the kinswomen of the groom, to build before the
marriage feast a new hut, in which after marriage they establish their
new home. Cf. Sudarabische Expedition, Bd. I, Die Somali- Sprache,
von Leo Reinisch, Wien, 1900, p. 107.

2 A Pilgrimage to Nejd, by Lady Anne Blunt, London, 1881, Vol. I,
p. 92.

8 Cf. Giddings's Principles of Sociology, p. 268.

56 SEMITIC ORIGINS

the mother, and that the women of the household were
the mothers and sisters, and not the wives and daughters,
of the men. The third of the possible arrangements of
Semitic marriage mentioned above turns out, therefore,
to be a true one.

Evidence is also at hand to prove that the fourth of the
possible arrangements was also realized in practice. In
three of the Mu'allakat poems there are specific statements
that the women whom the poets visited only occasionally
were members of other clans, and that often they visited
them at personal risk,i on account of the strained relations
of the clans. The marriage of Samson (Judges 14) was
also an alliance of this character. His wife resided in her
own clan, and he visited her there. In such cases as these
the marriages were often terminated by the migration of
the tribes in different directions.^ This is the general
type of marriage which Ammianus Marcellinus describes
when speaking of the Arabs, though he is probably speak-
ing of a somewhat later development of it. He says the
bride presents her husband with a spear and a tent, and
if she chooses withdraws after a certain day.^

This last phase of the marriage relation of the Semites
is probably but a modification of the heena marriage, or the
heena marriage a modification of it, brought about at times
by the hostile relation of the clans, as in the case of Sam-
son ; at times, by considerations of personal attachment to
his own clan, which made a man unwilling, even tempo-
rarily, to leave it; and at times, by economic necessities,
as will be pointed out below.

The general view which we have been led to take of
the marriage tie among the Semites is confirmed by the
position held among them by women in ancient times. So

1 See Mu'allakat of Labid, 11. 16-19 ; that of 'Antarah, 11. 5-11 ; and
that of Harith, 11. 1-9.

2 Mu'dllakat of Labid, 11. 16-19.

5 See Bk. XIV, ch. 4. He also remarks on the temporary character
of Arabic marriages.

PEIMITIVE SEMITIC SOCIAL LIFE 57

far from being the creature of man and almost his chattel,
as the system of selling daughters to become wives of the
baal marriages has made her, she occupied a position of
comparative dignity, equality, and independence. Smith
has shown that in Arabia, in pre-Islamic times, women
were frequently chosen as judges; that they were some-
times queens (of whom the queen of Sheba of Biblical
fame is best known) ; that they were regarded as the most
sacred trust of the tribe; and that, in spite of Moham-
med's humanitarian laws in behalf of women, their posi-
tion steadily declined under Islam in consequence of the
system of haal marriage, which practically made the hus-
band her lord.i This view is confirmed by Wellsted^
and Palgrave,^ who found that in Oman and Hasa, where
Islam is not so rigorously observed as in northern Arabia,
women were much more free and respected than in other
parts of the peninsula. In several places they did not
wear the veils even in the towns ; and in some, where it
was worn, the practice was voluntary.* This freedom is,
without doubt, a survival from pre-Islamic times.

In ancient Israel we also catch glimpses of a similar
freedom and dignity for women. In what appears to be
the oldest bit of literature in the Old Testament,* Debo-
rah figures as the inspirer and director of the people in
the movement for freedom. She assumes here a position
as free and prominent as any that woman occupied in
Arabia.

In Babylonia, too, the contract tablets reveal the fact
that at the close of the New Babylonian Empire, and in
the early Persian period, after many centuries of haal
marriage, women still held a position of great importance

» Cf. Smith's Kinship, pp. 100-106, 171, and 275.
" Cf. Wellsted's Travels in Arabia, Vol. I, pp. 351-354.
" Central and Eastern Arabia, Vol. II, p. 177.

4 WeUsted, op. cit., Vol. I, pp. 101, 118, and 146; also PalgraTe as in
n.4.

' The poem in Judges 5.

58 SEMITIC ORIGINS

and freedom. Married women appear with their husbands
as joint partners in buying, selling, borrowing, and loan-
ing; married women appear alone in business contracts
relating to money, real estate, and slaves ; they make con-
tracts concerning merchandise with men not their hus-
bands, and appear in lawsuits.^ This dignity, which the
Babylonian of the seventh and sixth centuries B.C. ac-
corded to woman, must be regarded as a survival of the
comparatively independent position which she held among
their early Semitic ancestors. Thus Arabia, Palestine,
and Babylonia each contribute to the proof of this position.
These arguments, taken in connection with the evidence
concerning the nature of primitive Semitic marriage, are
sufficient to make it clear that in the course of Semitic
progress the position of woman, in the family and in the
clan, has been greatly modified, and that she has lost in
the process much of her primitive importance. This point
will be still further confirmed when we come to consider,
in subsequent chapters,^ the religious argument. It will
then appear that in different parts of the Semitic territory,
notably in Arabia and Babylonia, goddesses survived till
a comparatively late time, who held a position of inde-
pendence of male deities, without parallel in later Semitic
social organization ; and whose birth would therefore com-
pel us, even if there were no other evidence on the matter,
to postulate a condition of society among the primitive
Semites in which woman should hold a position similar to
that described in the preceding pages. In many ways
free of restraint; often the head of her family, if not of
her clan ; usually leaving her maidenhood behind by one
or more acts of free love ; contracting marriages at will as
fancy dictated, but each of which was of short duration ;
cherished as the mother of her children and the perpetu-
ator of her family; performing the drudgery of nomadic

1 See the monograph of Victor Merx, " Die Stellung der Frauen in
Babylonien," etc., in Beitrdge zur Assyriologie, Vol. IV, pp. 1-72.
•^ Chs. Ill and VI.

PRIMITIVE SEMITIC SOCIAL LIFE 59

life, but mingling in, even when she did not direct, the
counsels of her uncles, brothers, and sons, — the primitive
Semitic woman was a picturesque figure, if not a model
for more modern days.

It is now time to inquire whether the primitive Semites
practised polyandry. Robertson Smith, following in the
footsteps of McLennan, interpreted many of the phenomena
we have passed in review as evidence of polyandrous prac-
tices. This, as is evident from the treatment accorded
the subject above, is not necessary. Such facts as we
have thus far examined may all be explained on the basis
of a temporary monogamy of the beena type, intermixed
with considerable sexual irregularity. Some of these
facts are not inconsistent, however, with the institution
of polyandry ; and there are others still to be considered
which make its presence at some periods and in some
localities certain.

Polyandry, in one form or another, has existed in many
parts of the world. It is found in India, both ancient and
modern, where it finds reflection in the ancient Maha-
bharata epic and other records ; and it still appears among
some existing tribes. ^ The most famous instance from
this land is that of the Nairs of the Malabar coast, whose
life has been most fully studied, and who represent one
type of the polyandric institution most completely.^ It is
also found in Thibet, though the kind of polyandry prac-
tised there is of another type.^ Still another type was,

' On polyandry in India, see Hopkins's monograph, "The Ruling Caste
in Ancient India," JAOS., Vol. XIII, pp. 170, 354 fi.; his Beligions of
India, Boston, 1895, pp. 467, 536 n. ; and his Great Epic of India, N.Y.,
1901, pp. 376, 399 ; Jolly's Becht und Sitte (in Buhler's Grundriss der
Indo-AHschen Philologie und AUertumsIcunde, Bd. II, Heft 8), pp.
47-49 ; Hilderbrand's Becht und Sitte auf den verschiedenen wirtschaft
lichen KuUurstufen, Jena, 1896, pp. 15, 16 ; Reclus's Primitive Folk, pp
143-177, and Starcke's Primitive Family, pp. 79-87.

2 Eor description see Reclus, as in n. 1. All writers on marriage and
sociology, from McLennan down, have much to say of them.

8 Cf. Spencer's Principles of Sociology, Vol. I, p. 659, and Starcke's
Primitive Family, p. 134.

60 SEMITIC ORIGINS

according to Csesar, found among the ancient Britons,^
and is still found among the Todas.^ Polyandry is also
found in the Polynesian Islands,^ until recently in Cey-
lon and New Zealand, in the Aleutian Islands, among the
Konyaks north of the Okhotsk, and among the Cossacks.
Humboldt observed it among the Indian tribes of the Ori-
noco; it was common in the Canary Isles; in Africa it
has been found among the Hottentots, the Demaras, and
among the mountain tribes of the Bantu race. It formerly
prevailed among the Picts and Irish.*

The explanations offered for polyandry are various.
McLennan believed that all races had passed through it
as a necessary stage on the way from promiscuity to
monogamy. Those who reject this view have assigned it
to different causes: some to poverty,^ others to natural
excess of males where tribes interbreed,^ and others regard
it as a mere incident of family communism.'' Poverty
cannot be the sole cause, since it is sometimes found
among the rich.^ It can hardly be explained by a natural
excess of males, since such excess is very improbable.^ As
a matter of fact, no one cause is sufficient to explain it in
all localities.^" Each instance of it must be studied by
itself in its peculiar environment and in the light of its
antecedents.

Before we return to Semitic polyandry it will be helpful
to glance at some of the different types of it which have
developed in different countries. These types are three

1 Cf . De Bello Oallico, V, 14, and Staroke, op. cit. , p. 139.

2 Cf . Spencer, op. cit. , Vol. I, p. 654.

5 Cf. Waitz, Anthropologie, Vol. VI, pp. 128, 129.

* MoLennan's Studies in Ancient History, p. 97 fif., and Giddings's
Principles of Sociology, p. 155 ff.

s So Hilderbrand, Becht und Sitte, etc., pp. 15, 16, and, in part, Gid-
dings, op. cit., pp. 155, 156, and 276.

' So Westermarok, History of Human Marriage, pp. 476-483.

7 So Starcke, Primitive Family, pp. 139, 140.

8 Cf. Westermarok, op. cit., pp. 476, 477, and 482.

' Cf. Starcke, In the International Journal of Ethics, Vol. Ill, p. 464.
w Cf. Spencer, op. cit.. Vol. I, p. 663.

PEIMITIVE SEMITIC SOCIAL LIFE 61

in number : Nair polyandry, in which a woman may have
as many as a dozen husbands, whom she receives in suc-
cession or as fancy dictates, but who in turn are free to
have as many mistresses as they can secure;^ Thibetan
polyandry, in which a group of brothers share one wife ;
and British polyandry, in which a group of sisters become
in common the wives of a group of brothers. These three
types represent three different forms of the institution.
Of these the Nair type is the most primitive. The Thi-
betan and British types may be considered as modifications
of the same form, since they are in principle the same.

Returning now to the Semites, we may note that the
type of temporary marriage, of which traces are found in
the Mu'allakat poems and in Ammianus Marcellinus,^ is
not necessarily monogamous or monandrous. In such
temporary unions, in which the husband and wife belonged
to different tribes, it would very probably be that each
would have acknowledged lovers in other tribes, with
whom they would have intimate relations whenever the
tribes approached one another so as to make it possible.
When we consider the sexual bent of the early Semites
and the lightness with which the marriage tie was regarded,
we can hardly hesitate to believe that this was so. Such
an arrangement might be classed as temporary monandry,
or as polyandry of the Nair type, according to the point
of view from which it is regarded. Like Nair polyandry,
it was at the same time polygamy. It differed, however,
from Nair polyandry in being exogamous; the Nairs
regarded intercourse with one of another caste as adul-
tery.^ In all probability, there was more polyandry than
polygamy in these marriages, for the practice of putting

1 Of. Eeclus, Primitive Folk, p. 163.

2 See above, p. 56. These temporary marriages, where the wife
received visits from her lovers with the ooirsent of her kinsmen, were
called mot' a marriages, i.e. marriages of pleasure or convenience. It is
given this name in some of the Arabic commentaries to Sura i^. See
"Wilken's Matriarchaat, p. 9, n. 3.

" Eeclus, ap. cil., p. 164.

62 SEMITIC ORIGINS

to death infant girls, which prevailed down to the time of
Mohammed (see Sura 16^i), was in all probability a primi-
tive Semitic practice. Where the conditions of life were
as hard as they always were in the Arabian peninsula,
more warriors than women would usually be needed by
the tribe ; and this mode of preventing not only too many
women, but too rapid an increase of the tribe, in view of
the limited means of sustenance, would be very natural.
Such a custom is not inconsistent with the high honor in
which the women who were permitted to live were held,
especially to a semi-savage mind not sensitive to incon-
gruities. Inevitably an excess of males would thus be
produced, which, among a sexually lax people, would be
sure to lead to polyandry. We cannot be sure that such
marriages, especially in later times, were always exoga-
mous. Those already cited from the Mu'allakat poems
certainly were, but there are others to be found in the
same collection which were endogamous. Imr-ul-Kais
alludes in his Mu'allakat to the fact that he followed the
women of his tribe and spent a day in their company, ^ and
the Unaizah, whose fruit he boasts he had repeatedly tasted,
was the daughter of his uncle. In like manner Laila, the
woman celebrated in the poem of Amr b. Kulthum, was
Amr's kinswoman. 2 In polyandrous marriages of the
general Nair type, there might exist both endogamous and
exogamous alliances ; and so far as this form of marriage
existed among the Semites, it would appear from extant
evidence to have combined the two kinds of marriage, that
from within and that from without the tribe. In so far
as Semitic feeling on this point can be historically traced,
it was in favor of endogamy ; Semitic parents were always
grieved if their children married outside their tribe. ^

1 See the Arabic commentator's explanation of v. 11 of Imr-ul-Kais's
Mu'allakat in Arnold's edition of the Mu'allakat. For a translation of
the poems, see The Seven Poems suspended in the Temple at Mecca, by
F. E. Johnson, London, 1894.

2 See Mu'allakat, Y, 11, 13, 14.

» See Genesis 243- ■•, 26^- 35, 28i- % Judges 14», etc.

PRIMITIVE SEMITIC SOCIAL LIFE

This is, however, probably a late feeling, which sprung
up when totemism was decaying, when primitive condi-
tions of marriage and kinship were breaking up, and when,
in disregard of earlier customs and ideas, the desire to
keep the children for one's tribe was gaining the ascen-
dency. In a totemic clan where real sisters are not taken
as wives, totemic sisters cannot be.^ As the Semitic
clans were totemic and did not, as a rule, marry sisters,^
we must infer that in the earlier stages of development
they were exogamous ; and that Nair polyandry, in so far
as it existed among them, existed as an exogamous insti-
tution. The mixed variety with which we meet in the
Mu'allakat poems is explained by the break-up of the old
religious ideas which was in progress, and the social
transition which the introduction of male kinship was
introducing.^

Perhaps the kind of marriage which is practised by the
Hassenyeh Arabs of the White Nile is a relic of the Nair
type of polyandry, though it might equally well be re-
garded as a slight limitation of the promiscuity of the
primitive Semitic girls, which we discussed above. Among
these Arabs, the marriages of the most respectable are not
for more than four days in the week, and may be for less
time. During these days the wife must observe the rules
for matrimonial chastity; but on other days she is free to
receive any man whom she may fancy, and the husbands
seem pleased with any attention paid to their wives during
their free and easy days, taking it as evidence that their
wives are attractive.*

It is safe to conclude, from the evidence presented, that
in early Semitic life a combined polyandry and polygamy,

1 See Giddings's Principles of Sociology, p. 271.

2 See Lev. 20i7- ", Qur'an 4^^ Yaqut, Vol. IV, p. 620, and Robertson
Smith's discussion, Kinship, p. 162 ff.

3 Wellhausen holds also that Arabic endogamy was preceded by exog-
amy ; see Nachrichten d. kgl. Gesell. d. Wiss. zu Gott., 1893, pp. 473 ft.

* See Wilken's Matriarckaat, p. 24, and Spencer's Principles of Soci-
ology, Vol. I, p. 617.

64 SEMITIC ORIGINS

approaching the Nair type, but originally exogamous, ex-
isted. It has, however, passed away, leaving few results
behind it which might not have been produced by a system
of temporary marriage, combined with a large degree of
that sexual laxity which exists among all peoples, in
greater or less degree, but which among the early Semites
was regarded as a religious duty.

Of the Thibetan type of polyandry we have more abun-
dant evidence. The most striking is the passage in Strabo's
description of Arabia Felix, often quoted by writers in
recent years : ^ " All the kindred have property in common,
the eldest being lord; all have one wife, and it is first
come first served, the man who enters to her leaving at
the door the stick which it is customary for every one to
carry; but the night she spends with the eldest. Hence,
all are brothers of all ; they also have conjugal intercourse
with mothers ; ^ an adulterer is punished with death ; an
adulterer is a man of another stock." This passage is
strong testimony of the existence in Yemen of fraternal
polyandry of the Thibetan type. It has recently been con-
firmed by the testimony of inscriptions brought from the
same region, Glaser stated, in 1897, that he had epi-
graphic evidence of polyandry, or communal marriage,
among the Sabseans,^ and Winckler, in the next year,
pointed out that in a Minsean inscription published by
Hal^vy, the genealogy demonstrated a fraternal polyandry.*
The evidence for this type of marriage for Yemen is there-
fore indisputable.

The late Robertson Smith collected considerable evi-
dence to show that this type of polyandry was also known

1 Strabo, Bk. XVI, ch. 4, p. 783.

'^ This is probably not to be taken literally, but to be explained by
Qur'au 4'^, where it appears that men had married wives of their fathers.
Cf. Robertson Smith in Journal of Philology, Vol. IX, p. 86, n. 2.

° See his note " Polyandrie oder Gesellschaftsohen bei den alten Saba-
em " in the Beilagen of Allgemeine Zeitung, Miinchen, December 6, 1897.

* " Die Polyandrie bei den Minaern," in Winckler's AUorientalische
Forschungen, 2te Reihe, Vol. I, pp. 81-83.

PRIMITIVE SEMITIC SOCIAL LIFE 66

in North Arabia and in other parts of the Semitic territory.
His arguments are: (1) Bokhari relates that two men
made a covenant of brotherhood, which resulted in their
sharing their goods and wives, — a fact which would seem
to show a survival of a custom of fraternal polyandry, i
(2) In Arabia karma means the wife of a son or brother,
but is used also to denote one's own wife. In Hebrew
kdlldh means both betrothed and daughter-in-law; while
in Syriac kalthd means both bride and daughter-in-law.
These facts can be explained most easily as remnants of
fraternal polyandry. ^ (3) The Arabic law that a man has
the first right to the hand of his cousin, as well as the fact
which the 4th Sura of the Qur'an and its attendant tra-
ditions attest, that in case a man died and left only female
children, the father's male relatives inherited his property
and married his daughters, are regarded as the results of
a previously existing polyandrous condition of society like
that described by Strabo.3 (4) The Qur'an (423) forbids
men to inherit women against their will, and forbids (4^8)
them to take their step-mothers in marriage " except what
has passed." This is regarded as evidence that down to
the time of Mohammed these attendant circumstances of
polyandry had continued, and that the prophet did not
dare to annul existing unions, though he forbade such
marriages in the future.*

The last two points quoted from Smith may not at first
sight seem to be valid arguments, but a little consid-
eration of the circumstances which would inevitably
attend polyandry of this sort, and the transition from
it to polygamy, will vindicate their character. In fraternal
polyandry the oldest brother is the head of the family,
and the wife is, or in time becomes, the property of the

1 Kinship, p. 135.

^ Ibid. , p. 136. I have modified the statements slightly in quoting
because, In the form in which Smith made them, they are not lexically
defensible.

» Ibid., pp. 138, 139.

*Ibid., pp. 86, 87.

F

66 SEMITIC ORIGINS

group of brothers. In case the oldest brother dies, the
next in age succeeds to his prerogatives and to his larger
claim on the wife. Thus, the idea is established that
inheritance carries with it not only rights of property, but
marital rights as well. Endogamous customs of marriage
extend this idea. A man comes to think of his paternal
cousin as by right his wife, so that the conception of in-
heriting women is strengthened and extends. Under this
system of polyandry the conception of male kinship grows
up and is firmly established, so that when polygamy suc-
ceeds polyandry the social soil is prepared for such cus-
toms as those urged by Smith as evidences of polyandry.
In this connection Smith also, following in the footsteps
of McLennan, urged that the Levirate custom of marrying
the wife of a dead brother to raise up seed to him, of which
we have such a beautiful idjd in Ruth 3, 4, of which he
also found traces in Arabia,^ and which still exists in
Abyssinia,^ was an outgrowth of fraternal polyandry. It
seemed to him and McLennan that no one would have
thought of counting the son of one brother as the son of
another, if previously the sons had not been the property
of all in common. Spencer, Starcke, and Westermarck
have all contested this position. Spencer suggests that
it is one of the results of inheriting women as one would
inherit other property;* to which Starcke justly replies
that this view leaves unexplained the real point of the
custom, the counting of the children as the offspring of
the dead brother. Starcke* and Westermarck^ point out
that the Levirate, or institutions of a similar character,
have existed in many parts of the world where there was
no suspicion of polyandry, and that therefore another
explanation must be sought. That which they offer is

1 Kinship, p. 87.

2 Letoumeau's Evolution of Marriage, p. 265.
' Principles of Sociology, Vol. I, p. 661.

* The Primitive Family, pp. 157, 168, and the International Journal
of Ethics, Vol. Ill, p. 465.

* The History of Human Marriage, pp. 610-514.

PRIMITIVE SEMITIC SOCIAL LIFE 67

that in primitive communities the idea of fatherhood is
juridical, and not based on actual fatherhood, and that
this fact, combined with the desire to keep intact the
dead man's estate, produced the institution in question.
The view of Letourneau,^ that the Levirate, though not
necessarily produced by polyandry, is practised under a
polyandric regime, seems to come nearer to the truth.
Possibly other customs and causes may have sometimes
produced an institution of a similar character; but when
we find absolutely certain evidence of the existence of
fraternal polyandry, such as we have for south Arabia, it
is but fair to interpret an institution which grows, as
we have seen, so naturally out of polyandry, as evi-
dence of its existence in another branch of the Semitic
race.

The explanations which Starcke and Westermarck give
of the Levirate seem inadequate in two respects : (1) They
leave unexplained why any one should desire to keep the
dead brother's estate intact, when it would be for the self-
interest of all the other brothers to have it divided; and
(2) they assume that in all parts of the world similar insti-
tutions must be produced by identical causes. Let it be
granted that polyandry does not offer a complete explana-
tion of why seed should be desired for an individual
brother, McLennan's contention that it did so is still so
far valid, that it may be said that polyandry supplies
some probable cause, while the juridical theory affords
none. It is not for an outsider to fight the battles of the
sociologists, but to me it seems more scientific to study
each institution in the light of its antecedents and envi-
ronment, than to heap instances together from every quar-
ter of the globe, and assume that because their external
character is similar one cause must have produced them
all. Studied in the light of the characteristics of the
Semitic race, we may still hold that for them the presence
of the Levirate system argues a previous polyandric con-
1 The Evolution of Marriage, p. 265.

SEMITIC ORIGINS

dition. This is the opinion of Wellhausen,^ Buhl,^ and
Benzinger,^ all of whom recognize that back of Arabic and
Hebrew life, as we know it, there lay a condition of
polyandry.

It is in this type of polyandry, where the wife is the
recipient of the favors of all the brothers, that the indi-
vidual father may not be known. It is always known that
a child is connected with a certain paternal stock, but
which one of the brothers begat him is a matter of doubt.
This led Robertson Smith to point out * that the Semitic
word abu must have originally meant "nourisher," not
procreator, and that in fraternal polyandry it must have
been applied to the elder brother. It thus acquired the
value of "husband" before it had the value of "father,"
and is actually employed in the former sense by Jere-
miah (ch. 3*). This observation led me to point out * that
in a Babylonian contract, which dates from more than two
thousand years B.C., the word ahu is also used in the sense
of "husband." This affords us at least one trace of this
system in Babylonia. It seems safe, therefore, to con-
clude that this type of polyandry began before the disper-
sion of the Semitic nations, or was developed by similar
circumstances, or was carried by later emigrants from
Arabia to the other nations. It seems probable that it
was developed before the later separations from the parent
stock occurred, and if not before the earliest, it was carried
to those countries by later migrations.

1 Cf. Nachrichten d. kgl. Gesell. d. Wiss. zu Gott., 1893, pp. 460 fi.,
474 ft., aiid479ft.

'^ Die sociale Verhdltnisse der Israeliten, von Franz Buhl, Berlin, 1899,
p. 28 ff.

^ Hebraische Archaeologie, Leipzig, 1894, p. 134.

* Kinship, pp. 117, 1.34. Cf. also my article, "The Kinship of Gods
and Men among the Early Semites," in JBL., Vol. XV, especially
p. 181 ff.

5 See my " Note on Meissner's AUbabylonisehe Privatreeht, No. 7," in
JA OS., Vol. XX, p. 326. The point of the article is that in line 24 of this
tablet a woman's father is called her abu, while in line 28 her husband is

PRIMITIVE SEMITIC SOCIAL LIEE 69

Of British polyandry, or communal marriage (the mar-
riage of a group of women to a group of men), there is
not, so far as I know, much evidence. Euting ^ describes
a caravan which he saw on its way from Haur^n to Kaf,
which contained 170 men and more than 20 young women.
This suggests the possibility that in the exigencies of
caravan life communal unions may have been formed.
Dozy 2 cites a case which occurred under Omar I, where an
old Arab gave to a young one a share in his wife, in return
for which the young man was to do gardening for him ; and
when reproved for it, both men professed to be ignorant
that they were acting contrary to law. This was, of course,
not communal marriage, but it indicates a point of view
which would make it possible, for convenience, to produce
such unions. If this type of marriage ever existed among
the Semites, it has left behind no sure traces of itself.

Having established the existence of Thibetan polyandry,
as well as that of the Nair type, we must inquire into
their relation to one another. Smith held^ that Thibetan
polyandry was a transition stage from the maternal to the
paternal family. As has been pointed out,* the Nair type
of polyandry is consistent with the conditions of very early
Semitic life, when marriage was exogamous. The type
of polyandry described by Strabo could only be introduced,
as endogamous Nair polyandry could be,^ when these con-
ditions were breaking up, when totemism was losing its
hold, and endogamy had taken the place of exogamy.
Smith also claims ® that the capture of women, of which
there is abundant evidence,^ had an important influence in

also called her abu, showing that the word was used in the same elastic
manner as it is in Jeremiah.

1 Tagbuch einer Seise in Inner- Arabien, Leiden, 1896, p. 38.

^ Histoire des Musselmans d'Espagne, par E. Dozy, Leiden, 1861,
Vol. I, p. 36.

8 Kinship, p. 144 ff. * See above, p. 57.

4 Above, p. 63 ff. " Kinship, pp. 74, 75.

' See also "Wellhausen, Nachrichten d. kgl. Oesell. d. Wiss. zu Gott.,
1893, p. 473.

70 SEMITIC ORIGINS

developing it. The sons of such women were, as Arabian
poets declare, brought up with their father's tribe. The
mother could not dismiss her husband at will, as in the
older mo^a marriage, but became subject to his power.
This power over her was sweet; and the advantage of
having their children to themselves, and not being com-
pelled to abandon them to the tribe of the mother, appealed
to them. But women were not always to be captured;
often, too, the conditions of life were too hard to allow of
the support of more than one in a whole family of brothers,
so the feeling against letting the children of sons go out
of the tribe would of course nurture the older feeling that
the children of the daughters were members of it ; and thus
gradually marriage with a kinswoman took, for the most
part, the place of extra-tribal, or clan marriages.

While the forces which transformed Nair polyandry into
that of the Thibetan type may have been, in part, those
which Smith supposed, there were other economic reasons
which, in Arabia, must have had a tendency to act in this
direction from the beginning. It is clear from Keasbey's
analysis of the clan organization, with which this chapter
opened, that the matriarchal clans, which we have in the
subsequent discussion proven to exist, must have had their
habitat in the oases of Arabia. There the women and
the weaker men would remain, thither other men would
from time to time repair, there Nair polyandry would be
practised, and there woman would be held in the high
esteem in which we have shown her to have been regarded
in Arabia. Such was the Arabic communal clan ; and to
it most of the evidence collected above applies.

From the beginning, however, there must have been a
tendency to the republican clan. Expeditions into the
desert with the flocks in search of pasturage, or caravans
from place to place for the purposes of trade, would con-
sist, as did the one which Euting saw, of a considerable
number of men and a much smaller number of women.
This would, from the beginning, have a tendency toward

PRIMITIVE SEMITIC SOCIAL LIFE 71

the formation of clans in which polyandry of the Thibetan
or British type would prevail. The women of the wealthy
Arabians of the oases who to-day accompany their hus-
bands on their expeditions into the desert are as a rule
those of lower social position. A princess in a harem may
have it understood that she is to remain always in the
oasis. 1 Probably it was so in ancient times. Such a band
of men would take with them some daring young women,
who had not much position at home, or who were captives
from another tribe. In such clans, where the men were
the most important element, and where Thibetan or British
polyandry would be almost certain, there would be a ten-
dency from the beginning to count the children to the
father's stock. The men of such clans, like some mod-
ern sailors, would be certain, too, to have mistresses in
every oasis which they visited; so that, while they formed
an important element in the social life of that Nair type
which we have traced above as the prevailing type among
the primitive Semites, they might also, in their own
migratory clans, have been laying the foundation of
Thibetan polyandry and paternal kinship.

The evidence passed in review goes to show that in the
most primitive times this tendency did not make itself
much felt. The reason why it did not is obvious. Arabia
is such a poor country, outside the oases, that the life of
the people is practically bound up in these fertile spots.
For a long time these adventurous bands were too depen-
dent upon the oases, and too much overshadowed by their
more numerous population, to make any marked impress
on the social order. As trade increased, however, and
the population, through numbers, was in places crowded
out permanently into the desert, such clans would become
more permanent; and thus clans practising Thibetan

^ Cf. Blunt, Pilgrimage to Nejd, Vol. I, p. 232. Only one of the three
■wives of the emir of Hail at the time of Lady Blunt's visit was bound to
accompany her lord on his expeditions into the desert. The other two,
•who never left the oasis, looked down on this one as an inferior.

72 SEMITIC ORIGINS

polyandry and counting kinship through the father might
be produced from economic causes.^ Wars would of course
be produced as a part of this process, so that marriage by
capture may have been one element of the transformation ;
but the economic element was probably earlier, and equally
prominent.

By these factors fraternal polyandry was produced.
Wellhausen^ ascribes to this feeling for one's tribe alone
the change from exogamy to endogamy. Fraternal poly-
andry adapts itself to a very poor country ; ^ and where
the murder of female children is added to the conditions
just described, it would seem to be the inevitable result
of the situation.

The restraint which this type of polyandry imposed on
men must always have been exceedingly irksome to those
who possessed the Semitic nature; and with them the
natural result would be that, whenever plenty permitted
the support of more women, and other circumstances threw
more of them into their power, polyandry would give place
to polygamy. This is what occurred whenever Semites
went into countries more fertile than Arabia. We find, as
we have seen, here and there traces of a previous polyandry ;
but wherever circumstances permitted it has given place to
polygamy, whether among Arabs, Babylonians, or Hebrews.

Hilderbrand, who has made a careful study of the family
type found among peoples who live by hunting, fishing, or
as shepherds and agriculturists, lays it down as a general
law that, " Among people who are in the lowest stages of
domestic development, we never and nowhere meet with
promiscuity or community in women."* This statement,

1 Such clans seem to exist in Arabia to-day and to have an organiza-
tion of their own. Samn, or melted butter, the produce of their flocks,
is their chief article of exchange. Ct. Doughty, Arabia Deserta, Vol. II,
pp. 71, 206-207, 209, 267, 268, 281, 289, 457.

2 Op. cit., p. 437 ft.

' Cf. Spencer's Principles of Sociology, Vol. I, p. 659.
* "Bei Volkern, welche sich noch auf der untersten wirtsohaftlichen
Stufe befinden, begegnen wir niemals und nirgens einem Zustande der

PRIMITIVE SEMITIC SOCIAL LIFE 73

I am told by good authorities in sociology, may be taken
as authoritative. He also tells us that, " The purchase of
wives is first found among peoples vp^ho have already
reached the condition of shepherd or agricultural life, and
individual property in land. Also, marriage by capture
is first frequently found in this stage of development."^
In like manner, with reference to polyandry, he deduces
from the cases he has observed this law, " Among peoples
who have already reached the shepherd or agricultural
stage of development and have individual property in
land, we not seldom find the phenomena that a number of
brothers or kinsmen possess one wife in common, or even
individuals live in complete celibacy." 2 Similarly, Gid-
dings remarks,' " The polyandrian family is found in very
many parts of the world, usually in tribes that have passed
beyond savagery into barbarism." ^

If, now, we apply the laws deduced by these students
of sociology to the ancient Semites, a part of the observa-
tions already made are confirmed, and, in some respects,
our knowledge of the earlier prehistoric period of Semitic
residence in Arabia is advanced. Hilderbrand's laws
confirm our view that the Thibetan type of polyandry is a
comparatively late development ; but they also lead us to
suspect that when the Semites separated from their Ham-
itic brethren of North Africa, they had already passed
beyond the lowest stages of social culture, since all our
data point to a sexual looseness for the primitive Semite

Frauengemeinscliaft oder Promiscuitat. " See Becht und Sitte, etc.,

p. 11.

1 "Der Sitte des Frauenkaufs begegnen wir emst bei Volkern, welche
schon auf der Stufe des Hirtenlebens oder aber des Aokerbaues und
Grundeigentums stehen. Und auch der Erauenraub kommt erst auf
diesen Stufen haufiger vor." Op. cit., p. 9.

2 "Erst bei Volkern, welche schon auf der Stufe des Hirtenlebens
Oder aber des Ackerbaues und Grundeigentums stehen, stossen wir nicht
selten auf die Erscheinung, dass haufig mehrere Brttder oder Verwandte
eine Frau gemelnsam besitzen, oder sogar Einzelne in einem Zustande
voUkommener Ehelosigkeit leben." Op. cit., p. 13.

3 Principles of Sociology, p. 155.

74 SEMITIC ORIGINS

which borders upon promiscuity. This observation is con-
firmed by two considerations : (1) There are certain fea-
tures in the Egyptian, as well as in the Semitic, religion,
which point to a previous condition of polyandry; ^ and
it is possible that the institution was developed before the
separation of the Semites from the Hamites. (2) The
conditions of life in Arabia and, to a certain extent, in
North Africa outside of Egypt, where, as in Arabia, there
are many deserts with occasional oases, are such that no
people could live long by hunting and fishing. The first
of these considerations will be more fully discussed in the
next chapter, but to the second some space may be devoted
here.

Fishing could never have been an important feature of
life in Arabia except upon the sea coasts, for the absence
of large rivers, and indeed, except in the oases, of water
of any sort, would render it impossible. Hunting has,
down to the present time, played some part in Arabian
life. Hares, wild goa'ts, gazelles, wild cows, and ostriches
may still be found in small numbers; and the Solluby
tribe, who have no real home, but pay tribute to all the
tribes, still live largely by hunting.^ If the theory of
Wallace, 3 that this region once contained larger forests
and more abundant water, be true, it can only have been
many, many centuries ago. Probably the camel and goat,
to which he ascribes the destruction of the forests, were
in Arabia before the Semites were. It is tolerably cer-
tain that, since the Semites entered it, the conditions of
the peninsula have been practically what they are to-day.
Here and there oases are found where a little water pro-
duces grass, trees, and vegetation, but in many of these
nothing of importance is produced without irrigation.*

1 Cf. Maspero's Dawn of Givilization, New York, 1897, p. 50 ff.

2 See Doughty's Arabia Deserta, Vol. I, pp. 281 ff., 362 ff., 487 ft,
Vol. II, pp. 9 ff., 70, and 216-218.

" See above, p. 26.

* On Arabian oases, cf. Vi^ellsted's Travels in Arabia, Vol. I, pp. 92 ff.
and 272 ff. ; Palgrave's Central and Eastern Arabia, Vol. I, pp. 20, 48 ff..

PRIMITIVE SEMITIC SOCIAL LIFE 75

Here and there, however, palms grow without artificial
watering.! Much of the country is covered with volcanic
mountains, from which protrude bare crags of igneous
rocks, and which produce almost no vegetation. The inter-
vening plains are covered with dry gravel, which is
exceedingly unproductive, while between the central and
eastern portions of the peninsula there extend immense
deserts of shifting sand.^ The lack of water and the
intense heat must have always made it difficult for savage
man to venture far from a spring. It is clear that in such
a country no large population could live by hunting; the
game itself would find the conditions of life too severe to
exist in large quantities. The Semite must have been
compelled to domesticate the goat and camel at an early
date, in order to obtain the milk which is so important a
part of Arabian diet. The date palm, which extended,
so Fischer and Hehn declare,^ in prehistoric times, from
the Canaries to Pen jab, and which now produces the staple
article of diet of so much of the Arabian population, must
have early revealed its virtues to the Semitic mind, and
thus called forth Semitic ingenuity for its cultivation.*

258 ff., Vol. II, p. 360 ; Blunt's Pilgrimage to Nejd, Vol. I, p. 113 ; and
Euting's Tagbuch einer Reise in Inner-Arabien, pp. 68, 121, 123 ff.

1 See Doughty's Arabia Deserta, Vol. II, p. 10, and Theobald Fischer
in Petermann's Mittheilungen, Erganzungsband XIV, No. 64, p. 10.

2 Cf. W^ellsted, op. cit., Vol. I, p. 241; Palgrave, op. cit., Vol. II,
pp. 132 fi., 136 ff., 153, 356-358 ; Blunt, op. cit., Vol. I, pp. 67, 156-185 ;
Doughty, op. at., Vol. I, pp. 419-422, 424, 425; and Euting, op. cit.,
p. 142 ff.

' See Theobald Fischer in Petermann's Mittheilungen, Erganzungs-
band XIV, No. 64, p. 1, and Hehn's Oulturpflanzen und Hausthiere, 6th
ed., p. 273.

* There should be no real doubt that the date-palm was known to the
primitive Semites in ancient Arabia. It extended in prehistoric times
from the Canaries to Penjab (see Hehn's CuUurpflanzen und Hausthiere,
6th ed., p. 273), or "from the Atlantic to the Himalayas" (so Theobald
Fischer, in Petermann's Mittheilungen, Erganzungsband, XIV, No. 64,
p. l),and "belonged to the desert and oasis peoples of the Semites"
(Hehn, op. cit., p. 263). This fact was doubted by von Kremer and Guidi,
as noted above in ch. i, on linguistic grounds, but without sufficient reason.

76 SEMITIC ORIGINS

Thus in Arabia, as has so often been the case in other
countries where the conditions of life are hard, necessity
compelled man at an early period to form a somewhat
advanced social organization. The conditions in which
such relations between the sexes as we have described

It is true the Semitic tongues have no common word for palm ; it is gis-
himmaru in Babylonian and Assyrian, diqld in Aramaic, tamar in Hebrew,
nakhlu" in Arabic, and tamrt in Ethiopic ; but as we pointed out above
(p. 22), Bertin has correctly observed (Journal of the Anthropological
Institute, Vol. XI, pp. 423-433), that it is the animals and plants which
are most common which always have the most names, and that some of
these may have survived in one dialect and others in others. It will be
noticed that the Hebrew and Ethiopic words for palm tree are identical.
Such a resemblance in two such widely separated dialects of the North
and South Semites shows, as Hommel long ago pointed out (Die Namen
der Saugthiere, p. 412), that this word was the name of it in the primitive
Semitic tongue. This is confirmed by the fact that in Arabic tamr means
" date," and then " fruit " in general, while tamara means to "feed with
dates." The use of tamr as date must have been a specialization of the
term for palm, when nakhlw^, the word for "tree," was narrowed to
mean "palm tree." That nakhlu", the more general term, could be
narrowed to the palm shows that that was the tree par excellence. The
Babylo-Assyrian term is apparently borrowed from a non-Semitic
people. Whence the Aramaic daqla came, it is not easy to say. Yaqut
(in his Geographical Dictionary, Vol. II, p. 580) speaks of a place, Dagala,
in south Arabia. " where date palms are found," which would show that
this term was also used in Sabsea. Perhaps it is this fact which led
Robertson Smith to say (^Religion of the Semites, 2d ed., p. 109), that
the date-palm was introduced into Arabia from Yemen and Syria, — a
statement impossible of proof. Surely the word daqld is not proof. One
could more plausibly prove from tamr that it was introduced from Pales-
tine and Ethiopia, which would surely be false. Hommel, when he wrote
Die Namen der Saugthiere, held that the date palm was a native of
Babylonia, but now says that it was introduced thither from Arabia
(Hastings's Dictionary of the Bible, Vol. I, p. 214). It is much more
likely, as Hehn says, that the palm was native throughout all North
Africa and Southwestern Asia. The culture of it would probably arise
first in an oasis country like Arabia, and may have been introduced
thence to Babylonia, as Hommel believes, and also to Egypt, as Hehn
thinks (op. cit., p. 274). Theobald Fischer, the scholar who has most
thoroughly investigated the date palm, holds that Arabia was the original
home of its culture, and it was thence introduced into Babylonia and
Egypt (op. cit; p. 11). The position taken in the text is therefore
thoroughly justified.

PKIMITIVE SEMITIC SOCIAL LIFE 77

could exist, even if, with Hilderbrand and Giddings, we
recognize that they can exist only in a pastoral and semi-
agricultural life, must have been present in the peninsula
not long, at most, after the Semitic occupation of the
country.

The importance of the date palm, for the sustenance and
development of Semitic life, can hardly be overestimated.
The palm leaves are to-day plaited into string mats and
baskets, and the bark into ropes. The dates themselves
form a staple article of Arabian diet, some of the people
having almost no other source of sustenance ; ^ they are
exported as far as Damascus and Baghdad,^ and in return
the Arabs are able to obtain a few articles from the out-
side world. The stones are ground and used for the food
of cows, sheep, and camels ; ^ syrup and vinegar are made
from old dates, and, by some who disregard the Qur'an,
a kind of brandy ; * and altogether the statement of Pal-
grave is not too strong : " They are the bread of the land,
the staff of life, and the staple of commerce." * They still
serve, in some parts of Arabia, as the standard of value,
as cattle do among shepherd peoples.® They cast a dense
shade, which, in contrast to the hot Arabian atmosphere,
must be exceeding ly_ grateful.^ Europeans regard the
dates as a not altogether pleasing staple of diet;^ but in
a land which produces so sparingly it is regarded as a
divine gift. An Arabic proverb declares that a good
housewife knows how to set before her husband a new

1 Cf. Doughty, Arabia Deserta, Vol. I, p. 148, Vol. II, p. 178.

' Central and Eastern Arabia, Vol. I, p. 60.

s In addition to the references in the two preceding notes, cf. Well-
sted's Travels in Arabia, Vol. I, pp. 94, 164 ff., 241, 288 ff., Vol. II,
pp. 112, 122, 419 ; Euting's Tagbuch einerlteise in Inner- Arabien, pp. 52,
53 ; Palgrave, op. cit.. Vol. I, p. 263 ; and Zwemer, Arabia, p. 123. For
the statement about vinegar and brandy, see Zwemer.

* Ibid.

' Central and Eastern Arabia, Vol. I, p. 60.

6 Doughty, op. cit., Vol. I, p. 332.

' Wellsted, op. cit. Vol. I, p. 94.

8 Palgrave, op. cit.. Vol. I, p. 60 ; and Doughty, op. cit, Vol. I, p. 148.

78 SEMITIC ORIGINS

preparation of date food each day in the month. ^ Much
thought has to be devoted to the culture of the date palm
in many places in order to make it grow. In many parts
of the peninsula it must be irrigated, and in some parts
water for the purpose must be conducted considerable dis-
tances. ^ The female flowers of the date palm must be
artificially impregnated from the male flowers, unless a
male tree happens to grow where the winds will naturally
carry the pollen to the female flowers. This is now some-
times done by planting a male tree in the midst of the
female ones ; but even as late as the early part of the pres-
ent century, Wellsted observed in the Sinaitic peninsula
an old method, once perhaps more widely used in Arabia,
of fastening a bunch of the male flowers on a branch ex-
posed to the wind, and so placed that it would disseminate
the jjollen over the flowers to be fertilized.^ In Mesopo-
tamia the method which the ancient sculptures attest, and
which is still employed,* was to climb the tree and sprinkle
the pollen over the flowers. This insured the fertilization
of each flower. That this tree and its culture played a
very important part in the development of ancient Semitic
life we may therefore well believe. Mohammed is said
to have addressed his followers thus : " Honor your pater-
nal aunt, the date palm. It was named our paternal aunt
because it was created of what was left from the clay of
Adam ; and it resembles mankind because it stands upright
in figure and height, and it distinguishes between its male
and female, and has the peculiarity (among plants) of
impregnating the latter. " ^ This high estimation of the

1 Erdekunde, von Carl Ritter, Berlin, 1779-1857, Vol. XIII, p. 804.
Cf. Zwemer's Arabia, p. 123.

2 Cf. Wellsted, op. cit., Vol. I, pp. 92-94 ; Euting, op. cit., pp. 52, 53 ;
and Glaser in Mittheilungen der vorderasiatische Qesellschaft, 1897,
pp. 373-376 and 425.

» Wellsted, op. cit., Vol. II, p. 12.
* Zwemer's Arabia, the Cradle of Islam, p. 123.

" Reported by Qazwini (1203-83, cf. Brockelmann's Qeschichte der
arabischen Literatur, Bd. I, Weimar, 1898, p. 481). The text is published

PRIMITIVE SEMITIC SOCIAL LIFE 79

palm is confirmed by an Aramaic inscription from Taima,
which, though much mutilated, shows that a part of the
fruit of a date orchard was consecrated to a god, ^ and by
the further fact that Nakhla, one of the seats of the wor-
ship of the goddess Al-Uzza,^ derived its name from the
date palm. The connection of the date palm with the
goddess will be established in the next chapter, and it
will there appear that the part played by this tree in the
evolution of Semitic civilization was of the greatest im-
portance. Fischer declares that the r81e which the Arabic
people have played in the world's history is closely bound
up with this, its sacred tree.^ If we substitute Semitic
people for Arabic, the statement remains equally true.
We can understand, from the economic value of this tree
and from the demand which its artificial propagation made
upon the Semite, as an increasing population made such
artificial culture necessary, something of the importance
it would assume in his eyes ; but to fully appreciate it,
we must learn the divine significance which he attached
to it, the reflex of his own social life which he saw in it,
and how he attributed to it all his knowledge, especially
the knowledge of sex and procreation. The social and
the religious life of the people are always interwoven.
These conceptions, which are so important for the social
life, as well as the religious feasts, which form so large
a part of the social intercourse of any people, will be
considered in the next chapter.

The discussion in the present chapter has, I think,
made the following points clear: The Semites, perhaps

in S. de Sacy's Chrestomathie arabe, "Vol. Ill, p. 175, French translation,
Vol. Ill, p. 395.

1 Cf. CIS., Ft. II, Vol. I, No. 113.

2 Cf. Wellhausen's Meste arabische Heidentums, 2(1 ed., p. 36; and
Sebraica, Vol. X, p. 64.

" "Wir konnen daher sagen, das auch die weltgeschichtliche RoUe,
•welche das arabische Volk gespielt hat, in engstem Zugammenhange mit
diesem seinem heiligen Baum steht." Petermann's Mittheilungen, Ergan-
zungsband XIV, Heft. 64, p. 10.

80 SEMITIC ORIGINS

as early as the time of their separation from the Hamites,
had reached the animistic stage of culture, and formed
totemistic clans. Their family relations were exceed-
ingly vague. Marriage was for a short term, women
resided in the homes of their own kindred, and descent
was reckoned through them ; the killing of female infants
created a paucity of women, which produced a condition
of polyandry resembling the Nair type. At the same
time there was much sexual irregularity, which was
regarded as innocent. Out of this there grew, through
the formation of small trading clans and the influence of
the capture of women, a system of Thibetan polyandry
and, later, a system of male kinship. Perhaps at the time
of their separation from the Hamites, and at all events
comparatively early, they had entered the pastoral and
semi-agricultural stage of culture, in which the cultivation
of the date palm played an important part.

CHAPTER III

SEMITIC BELIGIOUS OKIGINS

Many features of the religion of the primitive Semitic
people were successfully elucidated by the late Robertson
Smith in his epoch-making work, The Religion of the
Semites. In most respects it is, as yet, impossible to
advance beyond the position there taken. The primitive
Semitic community was, as he has so well shown, thought
by them to be made up of gods, men, and animals, all of
whom were akin to one another. All nature was peopled
with spirits, but the god of a people was the chief spirit
of the locality where that people dwelt. The gods were
confined each to its own tribe or clan, and in their activi-
ties they were limited to certain localities. They were
originally chthonic, and were identified with objects on
the earth before they were associated with heavenly bodies.
In this chthonic period they were especially associated
with springs, wells, and trees, and were regarded as the
proprietors of naturally watered land. The bond between
them and their worshippers was thought to be one of
physical kinship, and was believed to be renewed by sac-
rifice. The latter was originally conceived as a meal at
which both the gods and their worshippers partook of the
flesh of a victim which was akin to them both. Each clan
had its own god which it especially worshipped, though it
did not deny the reality of the gods of other clans. Each
god was limited in his activities largely to his own soil ;
and when one lived in the territory of a clan not his own
he must, in addition to his own god, worship the god of
the soil on which he resided.
G 81

82 SEMITIC ORIGINS

These positions Smith has satisfactorily established, and
it is not necessary to reopen their discussion here. In
one respect, however, it is possible to carry the investiga-
tion farther than Smith did, and to determine the gender
of the chief deities of the primitive Semites, the connec-
tion of their gods with the social organization outlined in
the preceding chapter, and some of the transformations
wrought in the conception of their nature by changed
economic conditions, migrations, and syncretism.

It is a law which may be regarded as practically univer-
sal, that the religious conceptions of a people are expressed
in forms which are modelled, in large degree, on those
political and social institutions which the economical con-
ditions of their situation have produced. Thus, a god
could not be conceived as a father where marriage was so
unstable that fatherhood was no recognized feature of the
social structure, nor as a king among a people into whose
experience the institution of kingship had never entered.
An illustration of this principle may be found in the fact
that republican institutions are, by their influence, gradu-
ally banishing the kingly idea of God from theological
discussions, and are leading to an emphasis of the father-
hood, and even brotherhood of God.i We should there-
fore, on general principles, be led to suppose that the
prominence of the mother and the institutions of mater-
nal kinship among the primitive Semitic clans, as well
as their tendency to unregulated intercourse and the
important functions of the date palm, all left a deep
impress on their religious ideas and practices. Indeed,
we may be sure that this is the case, especially as a
large mass of evidence has survived which is only intel-
ligible when interpreted in the light of these general
laws.

A considerable mass of this evidence was presented in

1 Cf . Can I believe in Ood the Father ? by "W. N. Clarke, Soribners,
1899, oh. iii; and "Fides et Spes Medici," by Dr. E. H. Thomas in
Present Day Papers, London, Vol. Ill (1900), p. 377.

SEMITIC RELIGIOUS OEIGINS 83

the writer's study, "The Semitic Ishtar Cult," ^ the main
conclusions of which are confirmed by further investiga-
tion. Additional material has also now been collected,
so that it is possible, in several respects, to carry the sub-
ject farther, and to prove more clearly than in 1894 that
the primitive Semitic religion was organized on the analo-
gies of its economic and social life. In the article men-
tioned the Ishtar cult was shown to be coextensive with
the Semitic peoples, traces of it appearing in Assyria,
Babylonia, north and south Arabia, Ethiopia, Nabathsea,
Moab, Palestine, Phoenicia, Cyprus, Malta, Sicily, and
Carthage. With three exceptions, the deity in all these
countries which received the largest share of the popular
homage was a mother goddess, and a patroness of unmar-
ried love. In Babylonia, Arabia, and Cyprus virgins
must sacrifice to her their chastity by an act of free love ;
at Byblos this might be commuted to a sacrifice of the
hair; and at Carthage and elsewhere her feasts were
attended by impure ceremonies, in which sexual excesses
formed a prominent feature. The Israelites found this
cult among the Canaanites, and adopted, as most scholars
hold, many features of its ritual. At all events, by the
time of the prophets the feasts of Yahwe were foul with
deeds most subversive of spiritual ideas.

Connected with this worship in historical times were
bands of priestesses (and often of priests) consecrated to
a service which, judged by modern standards, would be
prostitution. Ukhat, the creature who in the Eabani
episode ^ enticed that primitive man from his animals, was
a prototype and model of this order. With primitive
simplicity she unblushingly enticed him to the satisfac-
tion of desire, and is, in the Gilgamish epic, celebrated
for her act.^

1 Published in Hehraica, Vol. IX, pp. 1.31-165, and Vol. X, pp. 1-74.
Cf. also notes on the same topic in Hehraica, Vol. X, pp. 202-205.

2 See above, p. 43.

' Cf. Haupt's Nimrodepos, p. 11, 11. 16-21. For translations of the pas-
sage, cf. Jastrow, Religion of Babylonia and Assyria, p. 477 ; Jensen in

84 SEMITIC ORIGINS

There can be no doubt that the mother goddess whose
worship is thus widely diffused is a survival from primi-
tive Semitic times, when the mother held the chief place
in the clan, and all women shared a measure of free love.
As social conditions changed, the women who adhered to
the old practices would all have lost caste and become
despised harlots but for the fact that the social character
of the service of the goddess protected some of them. As
civilization advanced, it is probable that religious con-
servatism became a cloak for much that was vile and de-
basing. In the beginning, however, the practices which
were thus perpetuated must have been comparatively in-
nocent, since they but reflected the best thought of primi-
tive man with reference to manifestations of the divine.

The goddess Ishtar reflects, as was noted in the preced-
ing chapter, by her various unions the brevity of the mar-
riage tie among the primitive Semites. She married a
lion, a horse, and a bird, each for a brief space. She
desired to unite herself to Gilgamish, the hero of Uruk
(Erech), but he declined her advances. Thus, the myth
concerning her and the ritual by which she was served
reflect two different phases of primitive Semitic life, —
the temporary marriage and the consecration of the func-
tions of woman to the service of childbearing by one or
more acts of free love.

These features of her worship, taken in connection with
its universal diffusion among the Semites, renders us
certain of its existence in the primitive Semitic home.

It is important for us to note, also, that Ishtar was not
only the divinity who presided over human love, but over
all animal desire as well. Once when, according to an
ancient poem, she abandoned the earth for the lower world,
animals as well as men lost desire altogether. ^

KB., Vol. VI, p. 127 ; and Muss-Arnolt in Assyrian and Babylonian
Literature, Aldine ed. , p. 330.

1 Cf. Hebraica, Vol. IX, p. 147 ; and Jeremlas's Leben nach dem Tode,
p. 17.

SEMITIC RELIGIOUS ORIGINS 85

Connected with the worship of Ishtar was the worship
of the god called in Babylonia Dumu-zi or Tammuz. The
fourth month was named for him, and one of the chief
features of his worship was a ceremony of wailing for his
death, which was followed by wild rejoicing that he had
come to life. Prominent among the forms under which
this joy manifested itself was indulgence in unwedded
love.i Tammuz is variously represented in Semitic
mythology as son of Ishtar, as the first of her series of
rejected husbands, and as the beloved and lost husband of
her youth, whom she went to the under world to rescue.^
These myths represent conceptions which were formed by
three different stages of social progress. That which sees
in Tammuz Ishtar's son is a reflection of the primitive
Semitic family, the head of which is the mother, and the
chief male her son. The second, which makes him a
rejected husband, comes from a time a little later, when
marriage was still temporary and women quite free, but
when the original kindly relations between Ishtar and
Tammuz had been forgotten. According to this view, the
Tammuz wailing was a consequence of Ishtar's hatred and
vengeance, and not of her grief at his loss, as in the former
case. The third form of the myth reflects the later con-
ception of marriage as a more permanent and less sensual
relation. In the light of primitive Semitic social con-
ditions, there can be no doubt but that the first of these
conceptions is the original one.

Many scholars agree that Tammuz was in some way con-
nected with vegetation, and that the legend of his death
was a reflection of the annual dying of the leaves.^ To

1 Cf. Lucian's De Syria Dea, § 6 ; Ez. 8" ; and Hebraica, Vol. X,
pp. .31, 35, 73.

2Cf. II R. 36, 54; II R. 59, col. ii, 1. 9 ; IV R. 31, esp. col. li,
L 46 ff. ; Haupt's Nimrodepos, p. 44, 1. 46 fi. ; and Hebraica, Vol. X,
pp. 73, 74.

' Cf. Jensen, Kosmologie der Babylonier, pp. 197, 227 ; Frazer, Golden
Bough, Vol. I, pp. 278-396 ; Jeremias, Leben nach dem Tode, pp. 32, 41 ;
Nowack, ArchcBologie, Vol. II, p. 310 ; Bertholet, Das Buck Hezekiel,

86 SEMITIC ORIGINS

this opinion I adhered when the " Ishtar Cult " ^ was
written, and further study confirms it. Robertson Smith
was probably right in the opinion that the wailing at first
began as a mourning for the death of a theanthropic
victim,^ but there can be little doubt but that it was very
shortly associated with the death of vegetation, Lenor-
mant ^ and Halevy * are, I now think, wrong in claiming
a Semitic origin for the name, and it is not probable
that this origin was connected with vegetation except
indirectly.* Adonis, the name under which Lucian men-
tions him, is but an epithet which, in Phoenicia, had dis-
placed the original name, as other epithets displaced it
elsewhere. The original name is hopelessly lost.

The opinion expressed in the " Ishtar Cult " that Ishtar
was originally a water goddess, the divinity of some never
failing spring or springs, and that some sacred tree to
which the spring gave life represented her son,* can now
be confirmed by additional arguments.

The reasons which led to the adoption of that opinion
were : (1) that Athtar, in a number of Sabsean inscrip-
tions, is called " lord of the water supply " ; ® (2) an old

p. 49 ; Jastrow, Beligion of Babylonia and Assyria, p. 682 ff. ; and Toy,
Ezekiel in SBOT.,p. Ill ff.

1 Hebraica, Vol. X, pp. 73, 74. The hymn to Tammuz, IV K. 27,
No. 1, specifically connects 'him with vegetation. Cf. Ball's translation,
FSB A., Vol. XVI, p. 196. The name is, however, Sumerian, and means
" child of life " or " living child." It probably refers to Tammuz as the
child of the goddess of fertility.

' Beligion of the Semites, 2d ed., p. 411.

° Sur le nom Tammuz.

* Becherches bibliques, p. 95, and Melanges de critique et d'histoire,
p. 177.

' Hebraica, Vol. X, p. 73. I am now convinced, however, that the
name Tammuz is not primitive, but Sumerian Babylonian. It was at
times even applied to a goddess (see below, Chapter V). While the name
Tammuz was local (cf. my article, " The Genesis of the God Eshmun,"
in JAOS., Vol. XXI,2 p. 188 ff.), the god was, I believe, primitive, though
a less permanent and fundamental factor in the religion than the
goddess.

» See CIS., Pt. IV, Vol. I, No. 47, and Fell, in ZDMa., YoLlAY,
p. 245.

SEMITIC RELIGIOUS ORIGINS 87

Babylonian hymn calls Ishtar " the producer of verdure " ; ^

(3) the god Baal, with whom Ashtart in Phoenicia was
closely associated, was the god of well-watered land;

(4) the evident connection of Tammuz with vegetation;
and (5) the series of tree-like representations of the god-
dess found by Ohnefalsch-Richter in Cyprus.

To these arguments we may now add the following
considerations: the fact that in two inscriptions from
Gebal-Din, Athtar is the god of field fertility, which is
in Arabia especially connected with the water supply,^
forms another link connecting this cult with water and
vegetation. Ilmaqqahu, who, as is shown below,^ was
really Athtar under another name, was also the god of
field fertility.* Traces of tree-worship also appear, which,
if the Ishtar Cult represents the religion of the primitive
Semites, must be regarded as survivals from that time.
Trees were thought to be animate and to have perceptions
and passions, and were not infrequently taken as totems.®
In the latter case, all the attributes were ascribed to them
which under like circumstances were ascribed to sacred
animals. This proves the existence of that attitude of
mind on the part of the Semites which could easily see a
god in a tree. It still survives in Arabia, where certain
trees are thought to be inhabited by the jinn even to the ^
present time.® Such trees were probably in the pre-Islamic
days regarded as the residences of gods, who, upon the
introduction of Islam, shared the fate of other deities and
were deposed to the rank of evil spirits. In like manner
the Jews and early Christians regarded the gods of the
heathen as demons.^ Sometimes, however, it is not jinn

1 Cf. Zimmem's Bahylonische Busspsalmen, p. 33, and JSebraica, Vol;
X, p. 15.

2 See CIS., Pt. IV, Vol. I, Nos. 104, 105. » chapter IV.

* Cf. cm, Pt. IV, Vol. I, Nos. 72-102, and Fell, in ZDMO., Vol. LIV,
p. 244 ff.

6 Smith's Beligion of the Semites, 2d ed., p. 132.
8 Doughty's Arabia Deserta, Vol. I, p. 365.
' Cf, Deut. 32" and 1 Cor. lO^o.

88 SEMITIC ORIGINS

but angels who are thought to come down to tabernacle in
the trees ; and it is still the custom in parts of Arabia for
the sick to go to trees which are thus visited and offer
sacrifice and prayer for the recovery of their health. The
offering is usually a sheep or a goat, the blood is sprinkled,
the flesh cooked at the place, a part of it is divided among
the friends of the sick man and a part left hanging on the
branches of the tree. The worshipper then lies down and
sleeps, confident that the angels will come in vision and
speak precepts for his health so that he will rise whole.^
Such possessed trees are behung with old beads, votive
shreds of calico, lappets of colored stuffs and other such
things.^ This is a relic of old Arabian heathenism, in
which offerings were made in the same manner. The tra-
ditions tell that Mohammed referred to such a tree as " a
tree to hang things on." ^

Such traces of worship are not now found in connection
with the palm tree in Arabia, but more often with the acacia,
though at times with other trees and even with shrubs.
Some evidences of the worship of the palm tree in ancient
times are still extant. Tabari refers to the sacred date-
palm of Negran, where the tree was in all respects treated
as a god.* The residence of Al-Uzza at Nakhla, who was
in reality an Athtar,* is said by Ibn Abbas to have been a
group of Samura trees, in one of which the goddess espe-
cially dwelt. The Samura tree is explained by a scholion
to Ibn Hisham (p. 145) to be a palm tree.^ The reliabil-

1 Doughty's Arabia Deserta, Vol. I, p. 449 ff.

* Smith's Beligion of the Semites, 2d ed., p. 185.

' Cf. Annates quos scripsit at-Tabari, van M. J. de Goeje, Leyden,
1879-1897, Vol. I, p. 922, and Geschichte der Ferser und Araber zur Zeit
der Sassniden aus der arabisch Chronik des Tabari, von Th. Noldeke,
Leyden, 1879, p. 181. Smith (pp. cit., p. 185) holds that the statement
is incredible because it rests on the authority of a liar ; but liars some-
times tell the truth.

* Cf. Hebraica, Vol. X, pp. 58-66.

' Wellhausen's Beste arabische Heidentums, 2d ed., p. 38. Well-
hausen suspects this statement, because the vale of Nakhla (Palms) was
80 near. That, however, does not prove the statement wrong.

SEMITIC EELIGIOUS ORIGINS 89

ity of these statements has been unjastly suspected by
Wellhausen and Robertson Smith. The story of the birth
of Jesus, as told in the Qur'an, vouches for the ancient
sacredness of the palm. According to the statement of
Mohammed, which probably comes from Arabian Chris-
tians, Mary retired to a palm tree (Sura, 19^) as the time
of her delivery drew near, and was miraculously nourished
by dates produced out of season (19^^). Such a state-
ment reveals the conception that the palm tree was closely
related to the divine. All these references coincide with
a number of facts from other parts of the Semitic world
which indicate that the date palm was sacred, and thus
receive a confirmation which establishes a strong pre-
sumption of their truth.

In Abyssinia as in Egypt the sycamore was a sacred
tree, and in some instances still maintains this character.^

The terebinth was a sacred tree in Palestine. It plays
a prominent part in the traditions concerning Abraham
(Gen. 131^ 1413, 181), Gideon received a message from an
angel under one (Jud. G^^), and" in the days of Hosea in-
cense was burned under terebinths (Hos. 4^^). There are
traces also that the date-palm was a sacred tree in Israel.
Deborah is said to have sat under a palm tree, and is called
a prophetess (Jud. 4^), the inference being that the palm
was sacred, and that it helped her inspiration to be near
it. Some scholars endeavor to identify this with the tere-
binth of Gen. 35^ but without sufficient ground.^ There
is reason, as will appear below, to believe that the tree of
knowledge in Gen. 3 was a date palm. Evidence of this
also comes to us from the Jewish book of Enoch. In the
oldest portion of the Ethiopic Enoch we are told (ch. 24)
how Enoch visited paradise, and found that the tree of life
was a date-palm. 3 The full significance of this statement

1 Cf. Bent's Sacred City of the Ethiopians, p. 210.

2 So Moore, Judges in Inter. C'rit. Comm., p. 113, and Budde, Richter,
in Marti's Kurzer Sand Commentar, p. 35. On the other hand, cf. H. P.
Smith's Samuel in Inter. Grit. Comm. , p. 67.

8 Cf. Charles's The Book of Enoch, 1893. Charles rightly dates this
portion of the book before 170 b.c.

90 SEMITIC ORIGINS

will appear at a later point ; it is enough to note at pres-
ent that it affords evidence that the date-palm as a sacred
tree played a very important role in the thought of ancient
Israel. Other evidence of this is not wanting. The story
of Judah and Tamar (Gen. 38) indicates, since Tamar
means palm, that a clan was incorporated into the tribe
of Judah which made the palm a totem, and therefore
regarded it as a sacred tree. Further, on the confines of
Judah and Benjamin, there was a place, Baal-Tamar, which
took its name from a god who must have been called "lord
of the palm" (Jud. 20^). Earlier it seems to have been
called Baalat-Tamar, or " lady of the palm." In all prob-
ability the name was derived from an early connection of
a deity with a sacred tree.

At Elim, one of the stations at which the Israelites are
said to have stopped on the way out of Egypt, the palm had
a sacred significance, since it is connected with the sacred
number seventy and with twelve sacred wells (Ex. 15^).
Jericho, too, was called the " city of palm trees " (Deut.
34^, Jud. 11^ 3^3), and it is probable that there in early
times the palm had a sacred significance. The fact that
the palm tree and cherub formed part of the adornment of
the interior of the temple of Ezekiel (Ez. 41i^) and of the
temple of Solomon affords further proof of the same thing.
We cannot doubt, therefore, that the palm was a sacred
tree among the Hebrews or their immediate ancestors.

The numerous representations of trees on Babylonian
and Assyrian cylinders and monuments attest, as several
scholars have recognized, a primitive tree worship for the
ancient Babylonians or their ancestors. The tree most
represented, however, is the date-palm, and this is shown
to be in most instances the female date-palm by the hang-
ing clusters of dates. ^ Many of the representations on

1 Cf. Schrader in Monatsbericht d. kgl. prens. Ak. d. Wiss. zu Berlin,
1882, p. 426 ff. Other trees were also saored ; cf. Bonavia's article,
"Sacred Trees of the Assyrian Monuments" in the Babylonian and
Oriental Becord, Vol. III.

SEMITIC RELIGIOUS OEIGINS 91

the monuments picture a winged being, sometimes with a
human face and sometimes with an eagle's face, holding in
one hand a basket or bucket, and in the other a cone which
he is applying to the tree. That the difference in sex of
the date-palms was known to the ancient Assyrians is
attested by a fragment of a list of trees which was found
in the library of Assurbanipal, but which was probably
copied from a Babylonian list of much greater antiquity,
and in which gishimmaru zakiru, or " the male date-palm,"
is distinguished from gishimmaru zinnishtu, or " the female
date-palm."^ E. B. Tylor first suggested that the winged
figures which apply the mysterious cones to the trees are
representations of the winds — personified as divine agen-
cies — in carrying the pollen of the male flowers to the
stigmata of the female flowers, so as to fertilize them.^
He found in these figures the explanation of the cherubim
of Ezekiel and of Genesis, as well as of other parts of the
Old Testament. This seemed especially appropriate, since
Ps. 18^°, in a description of the coming of Yahwe on a
thunder cloud, equates the cherub with the wind. This
view has since been accepted by others,^ and affords a
most satisfactory explanation of these interesting repre-
sentations. Some of these portray a fish god, i.e. Ea in
the act of performing this fecundation. Ea was a water
god — a god of fertility, originally connected, as will be
shown by and by, with the primitive Semitic mother goddess.*
In the legend of Oannes, as preserved in Berossos, and
which is in reality a myth of Ea, a fish-like monster came,
we are told, from the sea, and taught the Babylonians the

1 Cf. II R., p. 46, No. 2, 11. 29, 30.

2 See FSB A., Vol. XII, pp. 383-393.

8 Cf. Jastrow's Religion of Babylonia and Assyria, p. 662, and Haupt
in Toy's Ezekiel in SBOT., pp. 181-184.

* See Chapter V. The fertilization of the date-palm in Mesopotamia has
to be performed in part by hand nnto the present time. (See Zwemer's
Arabia, the Cradle of Islam, p. 123.) This fact explains the anthropomor-
phic form of the cherub. The wind is conceived as a supernatural man
applying the fertilization by hand.

92 SEMITIC OEIGINS

beginnings of civilization. Among other arts he made
them distinguish seeds, and taught them how to collect
fruit. In his hands, therefore, the cone and bucket would
properly have a place.

The fact that the sacred character of trees is established
for so many parts of the Semitic area is good evidence that
tree worship existed in the primitive Semitic home, and
the traces of the sacred character of the date-palm which
have been adduced above lead us to think, when taken in
connection with what we learned in the previous chapter
of that tree, that it was the sacred tree par excellence.

In the same way the idea that perennial springs and
wells were connected with divinities is found in many
parts of the Semitic territory, and is no doubt primi-
tive. At Beersheba, Dan, Sidon,^ on Mount Lebanon,^
and at Mecca sacred wells and springs were found, not
to mention many others. That they do not appear in
Babylonia is due to the presence of the great rivers and the
nature of the irrigation of the country ; but the fact that
Ea was a water god attests there the same circle of ideas.
We cannot therefore be far wrong in coupling the two —
the palm tree and the spring — and in seeing a mythologi-
cal representation of them in the primitive mother goddess,
Ishtar, and her son Tammuz. Indeed, the well at or near
Sidon was sacred to Eshmun, a god who, as I have pointed
out elsewhere,^ was probably developed out of Tammuz by
the use of an epithet.

That this is a correct view is confirmed from another
quarter. The Assyriologists of a score of years ago * re-
garded the pictures of the sacred tree on Babylonian and

1 CIS., Pt. I, Vol. I, No. 3, 1. 17.

2 Cf. Hebraica, Vol. X, pp. 33 and 61-66, and Smith's Beligion of the
Semites, 2d ed., pp. 166 ff. and 177 f.

" See the article " Ashmun " (Eshmun) in the Jewish Encyclopedia, and
"The Genesis of the God Eshmun" in PA08., Vol. XXI 2, p. 188 ff., and
below, Chapter VI.

* So George Smith in his Chaldaean Genesis ; see the German transla-
tion, p. 84 ; also Lenormant in his Origines de Vhistoire, Vol. I, p. 90.

SEMITIC RELIGIOUS ORIGINS 93

Assyrian monuments as the prototype of tlie tree of life in
Gen. 2 and 3, and though this view is rejected by some ^
who see in these pictures the date-palm, it is, I believe,
a hint in the right direction. The Yahwistic writer of
Gen. 2 and 3 gives us a twofold representation of the cir-
cumstances of the union of Adam and Eve and its effect.
In ch. 2 we are told how man, after consorting with the
beasts, left them for the woman and became "one flesh"
with her. That this was the original form of the story the
parallelism of the Eabani and Ukhat episode in the Gil-
gamish epic enables us to determine.^ It also enables us
to see that the Rabbis were right in explaining " cleave to
his wife and become one flesh " as referring to connubial
intercourse.^

In ch. 3 the same thing is differently represented. A
serpent tells the woman to pluck the fruit from a for-
bidden tree, she does it, the man and woman both eat of
it, their eyes are opened, and they know good and evil.
The first effect of this knowledge was the perception of
the difference of sex — the perception that they were
naked. The Rabbis thought that the serpent here repre-
sented the sexual passion,* but it is doubtful whether
this is correct. As among many other peoples, the ser-
pent was sacred ^ among practically all the Semites. He
belonged no doubt to that primitive totemistic circle of
society of which we have seen the primitive Semitic com-
munity to consist. He is here represented as living in
the primitive garden or oasis, where serpents no doubt
abounded, and as urging man to partake of an act
which seemed to him a divinely provided joy. In ch. 2
and its Babylonian parallel, woman enticed man from in-

1 So Bonavia, Babylonian and Oriental Record, Vol. Ill, p. 36 ff,

2 Cf. Jastrow's "Adam and Eve in Babylonian Literature " in AJSL.,
Vol. XV, pp. 192-214.

3Cf. Jastrow, AJ8L., Vol. XV, p. 207, n. 44.
M^5'i., Vol. XV, p.209.

« Cf. Pietschmann's Geschichte der Phonizier, p. 227, andKitteV sKonige
in Nowack's Hand-Kommentar, p. 278 ff.

94 SEMITIC ORIGINS

tercourse with the beasts, so here a beast is represented as
urging man to union with woman. The two representa-
tions arose no doubt because of the union of two originally-
independent explanations. The effect of tasting this
divine fruit was that man was thereby brought to the
knowledge of good and evil, i.e. to the exercise of a virile
manhood ; he was led to adopt clothing, to till the soil, and
to a knowledge of the various features of civilization. This
view of the meaning of good and evil is confirmed by
the fact that in Deut. 1^ " having no knowledge of good
and evil " is equivalent to not having reached the age of
puberty.

But why in this case should the tree appear at all?
Why should its fruit even symbolically represent such an
act ? The answer is, I believe, to be found in the fact that
the beginnings of Seinitic civilization were connected with
the date-palm, that a knowledge of the difference of sex
in these trees was known at a very early time, and that
the marvellous effect on the palms of the fertilization
wrought by the wind, appeared to the primitive Semitic
mind as a divine exhibition of sexual fertilization and
divine approval of it. Thus, the two would become asso-
ciated in the Semitic mind, and in time the act would
naturally be pictured as the fruit of the tree. That this
view represents the truth is shown by the fact that in the
Biblical narrative cherubim are placed in the gate of the
garden to prevent the return of man to his Eden of sexual
unconsciousness. The cherubim were, as we have seen,
representations of the wind, which bore the fructifying
pollen of the male flowers to the female, and the introduc-
tion of the cherubim at this point represents the primitive
feeling, that the constant enaction of this divine process
of fertilization in the tree, which stood in the garden of
his god and which sustained life, forced man onward by its
divine example to similar acts with all their consequences.
The ever-present cherubim of the palm kept alive, he
thought, the sexual passion in himself which made absti-

SEMITIC RELIGIOUS ORIGINS 95

nence and a return to conditions which he regarded as
primitive (i.e. a life in which woman played no part)
impossible.^

Thus the Assyrio-Babylonian sacred tree becomes not
the prototype in the first instance of the tree of life, but of
the tree of knowledge of good and evil. A comparison
of Gen. 33 -^ with Gen. 2^ and 3^2 reveals the fact that in
the original form of the story only one tree is mentioned.
The tree of life in the two latter verses is a later addi-
tion. ^ That such additions should be made to the sub-
structure we have supposed is shown to have been very
natural by the following facts. The idea of a future life
played no important part in primitive Semitic thought.
The life of the spirit after death was thought by the Baby-
lonians and Hebrews to be a colorless and undesirable one,
and to the Arabs of the desert, the idea of an under world,
seems to have been wholly lacking.* The problem which
confronted them was the cause of present suffering, and
not the problem of an immortal life. As the thirst for an
immortal life was felt, but before it had been accepted as
a fact, the story of the cause of human suffering would
naturally be modified to make it explain why man could not
live forever, and this is the form which we have in Genesis.
As time went on and a provisional immortality of five hun-

1 Thus an Arabic poet describes and addresses the palm : —

"He lifts his leaves in the sunbeam glance
As the Almehs lift their arms in dance ;
A slumbersome motion, a passionate sigh,
That works in the cells of the blood like wine.
O tree of love, by that love of thine
Teach me ho w I shall soften mine. ' '
— Translated by Zwemer, Arabia, the Cradle of Islam, p. 121 ff.

" See Die Biblische Urgeschichte, von Karl Budde, Giessen, 1883, p.
53 ff., and Toy, JBL., Vol. X, p. 12 ff.

3 See Der AhnenkuUus und die Urreligion Israels, von Carl Griineisen,
Halle, 1900, p. 55. This seems true notwithstanding Stade {Geschichte,
Vol. I, p. 395, n. 2) , and Sohwally (Leben nach dem Tode, p. 46, etc.). For
Babylonian view see Jeremias, Leben nach dem Tode, and for the Hebrew,
Charles, Eschatology, p. 33 ff.

96 SEMITIC ORIGINS

dred years was accepted (Eth. Enoch, 10^"), the tree af
knowledge disappeared from Eden and the tree of life t#)k
its place (Eth. Enoch, 24, 25). Thus did Hebrew thought
transfer the story from an explanation of toil to the prom-
ise of future reward.^ This transfer was easy; for, in
another sense, the tree was to the primitive Semite always
a tree of life as well as a tree of knowledge. The parallels
which the Eabani story affords to the narrative of Genesis
vouch for the Babylonian derivation of the latter. This
is also shown in the fact that the garden is situated in the
East (Gen. 2^), and that the Tigris and Euphrates rivers
are mentioned in connection with it^ (Gen. 2"-i^).

There are in the Eabani episode, as has been already
pointed out, features which were derived from the primi-
tive conditions of Semitic social life. Although these
features have been somewhat veiled in the Biblical narra-
tive, they are nevertheless present, and that narrative also
contains another primitive feature which is still more
prominent. The narrative in Gen. 3 represents God,
man, and the serpent as forming one social circle. The

1 On the view presented in the text the historical origin of the Hebre-w
ideas of Eden and the heavenly paradise or New Jerusalem are as fol-
lows: The primitive conceptions of a sacred enclosure, where the god
dwelt and the sacred tree was, grew out of an Arabian oasis, or possibly a
North African, at a still earlier time (see below). This was transferred to
Babylonia, where it became a garden. This conception was taken over
by the Hebrews and is represented iji Gen. 2 and 3. As time passed on
and Jerusalem was destroyed and rebuilt, the Jewish ideal passed from
a garden to a city. A garden may have been the home in the beginning,
but a*ity became their ideal for the future. As Apocalypses were writ-
ten and their authors sought for imagery under which to shadow forth
their hopes of the heavenly future, they sometimes took the picture of
Eden as did the author of Eth. Enoch, 24, 25 ; sometimes the city of Jeru-
salem, as did the author of Psal. Sol. 17 ; and sometimes the two were com-
bined as in the Apocalypse of John, where it is a city with twelve gates
(ch. 21), and yet it has a river with a tree of life, i.e. a garden (oh. 22i-^).
Thus the imagery born in prehistoric times in the Arabian oasis with its
palm tree appears, transformed and elevated it is true, but still appears on
the last page of the New Testament.

^ See Delitzsoh's Wo Lag das Paradies, Leipzig, 1881, and Haupt,
in Ueber Land und Meer, 1894-95, No. 15.

SEMITIC RELIGIOUS ORIGINS 97

serpent is wiser than man ; he talks to the woman, and his
power of speech causes her no astonishment. These ele-
ments of the tale must have taken shape in a primitive
totemistic society in which animals were really believed
to possess such powers ; i.e. it reflects the conditions of
primitive Semitic, and not of Hebrew thought.

The fact that in Genesis Yahwe is represented as for-
bidding the acquisition of the knowledge of good and evil
on the part of man, has, I think, nothing to do with the
primitive form of the story; it is but the local coloring
given to the tale by the Yahwistic writer. This writer, in
the stories of Cain and his descendants which follow,
attributes the beginnings of civilization in every instance
to those who disobeyed Yahwe. When he does so in the
narrative of Eden, he is but following out his prevailing
tendency. An opportunity was afforded him to thus
interpret the tale by one of the features of the Babylonian
story, preserved on a fragmentary tablet, which may form
a part of the Gilgamish epic. This represents Eabani as
cursing Ukhat, who had promised to make him like a god
and who had instead brought him to death.^ This story
probably reflects the evil effects of the unrestrained sexual
practices of the Semites, as does also that passage in the
sixth tablet of the epic where Ishtar's love is represented
as so terrible that she has smitten and crippled all her
husbands. Such loose sexual habits as those traced above
would necessarily produce venereal disease and death, and
such dire effects might well be interpreted as evidences of
the anger of the god. In the Eabani episode this view is,
so far as we can tell, not taken. Eabani's anger is directed
against the woman alone ; he does not seem to be conscious
that he has angered a god. This latter inference, how-
ever, lay close at hand, and could hardly fail to be made
by a writer whose attitude toward civilization was like

1 Cf Haupt, JVmrodepos, pp. 16, 17, and 5A, Vol. I, pp. 318, 319;
also Jastrow, Religion of Babylonia and Assyria, p. 478, and AJ8L.,
Vol. XV, p. 209.

98 SEMITIC ORIGINS

that of the Yahwist. Side by side with the Babylonian
view just described, and older than it, was another, which
attributed civilization to the knowledge of sex and which
regarded both as a blessing. Divine approval was mani-
fested through the example of the sacred tree, which
was the home of the divinity. It is thus only that we
can account for the reference of civilization to sexual
relations, for the sacred character attached to those rela-
tions among the Semites, and for the connection of both
with the sacred tree.^

Thus our view of the original form of Ishtar is con-
firmed since in the palm tree, which grows by every
Arabian spring, and which has grown there since man
inhabitated Arabia,^ we find that the Semite saw the em-
bodiment of all those features of vegetable and animal
fertility which characterize this primitive Semitic cult,
and which found such expression at a later time in its
religious practices and in its mythology. Since we are led
by such reasons to these conclusions, it seems most natural
to find in the rite of circumcision, which has survived
among the Arabs, Abyssinians, Syrians, Phoenicians, and
Hebrews, a confirmation of them, and in them an explana-
tion of Semitic circumcision. Circumcision has been
found among many peoples of the world, and is usually
explained like tattooing, cutting off a finger joint, and
other mutilations, as embracing the twofold idea of offer-
ing a sacrifice to the god and furnishing a tribal mark by
which the god may easily know his followers, and they
may be known to each other. That it had this latter
force among the Semites is attested by its history among
the Hebrews. The Yahwistic writer represents Yahwe

1 The Biblical writer is in this representation also paralleled by
another Babylonian tale, the Adapa myth (of. KB., Vol. VI, pp. 92-
101). This myth represents the god Ea as preventing hy a deception the
eating of the bread and water of life (i.e. the gaining of immortality), hy
a mortal. Cf. Gunkel, Schoffing und Chaos, p. 148 ff., and Jastrow, Be-
ligioH of Babylonia and Assyria, p. 549 ff.

^ See Theobald Fischer in Petermann's Mittheilungen, Erganzunungs-
band XIV, No. 64, p. 11, and above. Chapter II, p. 75, n. 4.

SEMITIC RELIGIOUS ORIGINS

(Ex. 4^-^) as trying to kill Moses or his son as though
he were of a foreign stock untU Ger shorn was circumcised,
when Yah we desisted ; while the priestly writer regarded
circumcision as the sign of Yahwe's covenant with his
people (Gen. 17'°"^^ Ex. 12**). Such passages attest the
religious importance of the rite among the Israelites, and
the struggle which Paul and the early Christians who
thought like him were compelled to undertake to gain
emancipation is sufficient, to mention no other evidence,
to show the importance attached to it hy the Jews as the
visible sign to their god and to one another of their
fidelity. 1 Herodotus mentions the Syrians and Phoeni-
cians among those who practise circumcision, ^ but of the
details of its practice among them we know nothing. Of
its practice among the ancient Arabs we have fuller infor-
mation. It is mentioned by Josephus and Sozomen as a
practice of the northern Arabs, and by Philostorgius as a
practice of the Sabseans.^ Sharastani mentions it as one
of the practices which Islam confirmed as a religious duty.*
The way in which it is observed in Arabia at the present
time attests the truth of this statement. Among the
Bedawi it is the occasion of a feast at which the rite is
performed on children of three full years. There is danc-
ing on the part of the maidens, while the young men stand
about and select from the dancing throng their wives. A
sheep is sacrificed, its flesh cooked and eaten near sun-
down at a feast, while the entrails are left hanging on a
trophy bush, or sacred tree. After the feast the dancing
begins again and continues into the evening.^ Among

1 For a concise sketch of the history of the rite in Israel see the article
"circumcision" by McAllister in Hastings's Dictionary of the Bible, or
by Benzinger in the Encyclopedia BiUica. For the Abyssinian custom,
of. Wylde's Modern Abyssinian, p. 161.

2 Bk. II, ch. 104.

8 Josephus, Ant. I, 12^ ; Sozomen, H. E., VI, 38 ; Philostorgius, H. E.,
Ill, 4. Of. also Nowack's Archdologie, Vol. I, p. 167.

* See Haarbruoker's translation of Sharastani, Vol. II, p. 354.
s Doughty 's Arabia Deserta, Vol. I, pp. 340, 341.

100 SEMITIC ORIGINS

other Arabs it is the custom to make the child ride on the
back of the sacrificial sheep. ^ At Mecca there still exists
a similar custom of performing circumcision in connection
with a sacrificial feast.^ Here the operation is performed
from the third to the seventh year, and is performed on
female children as well as upon male.

The circumstances under which it is performed in
Arabia point to the origin of circumcision as a sacrifice to
the goddess of fertility, by which the child was placed
under her protection and its reproductive powers conse-
crated to her service. The slaughter of the sheep was
originally not simply for domestic purposes, since all
slaughter of domestic animals was sacrificial.^ The con-
secration of the child by such an offering, in addition to
the regular sacrificial victim, is parallel to the sacrifice of
chastity by which women consecrated their wombs to the
goddess of childbearing at Babylonia and Byblos.* In the
dance and the selection of future wives by the young Arabs
in the Bedawi ritual we see a survival in a purified form
of an old love feast, such as must have accompanied in one
form or another all the feasts of the Semitic mother god-
dess, and to which Augustine and Ephraem bear witness.^
Originally circumcision seems to have been a preparation
for connubium.^ Its transfer to infancy may, as W. R.

1 Doughty, op. cit., Vol. I, p. 391.

2 See Snouck Hurgronje's Mekka, Vol. II, pp. 141-143. Among the
Hamitic Somalis of East Africa, who are deeply penetrated with Arabic
influence, boys are circumcised at seven years of age, and girls are infibu-
lated at ten. The hair is cut short at the same time, so that a long-haired
person and an uncircumcised are identical. Cf. Reinisch, Somali-
Sprache, pp. 110, 111, etc. ; Bd. I of his SUdarabische Expedition. Wien,
1900.

3 See Smith's Meligion of the Semites, 2d ed., pp. 234, 241, and 307.

* Herodotus, I, 199, and Lucian's Be Syria Dea, § 6. Cf. Hebraica,
Vol. X, pp. 21 and 31.

' Cf. Ephraem, Opera, Vol. II, pp. 458, 459 ; Augustine, De Civitate
Dei, II, 4 ; also Hebraica, Vol. X, pp. 51 and 59.

^ Cf. Gen. 34, and also Ex. 4^6, where circumcision is connected with
the idea of " bridegroom."

SEMITIC RELIGIOUS ORIGINS 101

Smith suggests,^ have been a later development. Circum-
cision thus receives for the Semitic peoples a fitting
explanation, and an explanation not out of harmony with
that usually given it by modern scholars for other peoples.

Circumcision was also practised by the Egyptians at a
very early date,^ and Herodotus was so impressed by their
practice of it that he claims that others learned it from
them. 2 According to Strabo they, like the modern
Meccans, circumcised both men and women.* The Gallas,
another Hamitic tribe, also practise it.^ The connection
of this Hamitic practice with the Semitic will be consid-
ered with some other similar matters at a future point in
the discussion.

In a system of religious thought, in which the sexual
functions of the animal world found a counterpart and an
apotheosis in the processes of the sacred tree, and in which
free love was at certain times a religious duty, what more
natural than that the organs of reproduction should be
placed under the care of the tutelary divinity by such a
sacrifice ? Indeed, the Arabs to-day, who are much with
flocks and herds, declare that only in man is an impediment
like the foreskin found, and wonder how it is possible for
reproduction to occur among uncircumcised Christians.*
Possibly their remote Semitic ancestors reasoned in the
same way, and so conceived the necessity of making this
sacrifice to the goddess of productivity, that they as well
as other creatures might receive the blessing of fertility.

Trumbull has collected a convincing array of instances
of the sacred character of the threshold among the Baby-
lonians, Phoenicians, Hebrews, and Arabs,^ which prove

1 Cf. Bel. of Sem., 2 ed., p. 328; "Wellhausen, Seidentum, 2 ed.,
p. 175.

' Ebers, Aegypten und die Bucher Moses, Vol. I, p. 283.

8 Bk. 11, 104.

4 Strabo, Bk. XVII, 25.

-" Maoalister in Hastings's Dictionary of the Bible, Vol. I, p. 444.

« Doughty's Arabia Deserta, Vol. I, pp. 341 and 410.

' The Threshold Covenant, Philadelphia, 1896, pp. 108-164.

102 SEMITIC ORIGINS

that the threshold among the Semites, as among people in
many parts of the world, had the sanctity of an altar.
The explanation which Trumbull offers for the sacredness
of the threshold throughout the world is that primitive
men everywhere make, by some common psychological
process, a connection between the relation between the
threshold and doorpost on the one hand, and the relation
of the sexes on the other. ^ The result of our investigation
into Semitic religious origins confirms this conclusion in so
far as it applies to the Semites. A people who, like them,
attributed to the sexual relation the beginnings of intelli-
gent life, the knowledge of clothing, agriculture, and the
arts of civilization, and who, in their conceptions of
divinity, in their religious rites, and in their social organ-
ization, gave such prominence to sexual relations and
functions, would most naturally invest the threshold, the
approach to the tent or house where the fruits of these
divinely ordained functions were sheltered, with something
of the sanctity of the function itself. This would be
especially easy for early man as soon as any structure
beyond a mere tent formed his dwelling. The old Semitic
door sockets and posts would by their very form readily
suggest the organs of fertility. No doubt the nosb or
masseha, which bore a general resemblance to a phallus,
afterward became the symbol of Semitic deity for a similar
reason.

With the view of the nature of Ishtar here set forth,
the meaning of her name, I believe, coincides. It is true
that so many varying etymologies of the name have been
offered that it is precarious to build much on an argument
derived from this source, and yet the confirmation afforded
to the views expressed above by what I believe to be the
true etymology renders it worthy of a little discussion.

Driver 2 still refers the name, as Schrader^ and Sayce*

1 Op. cit., pp. 193-203.

2 " Ashtoreth " in Hastings's Dictionary of the Bible.

3 KAT.% p. 179. i Hihbert Lectures for 1887, p. 252 ff.

SEMITIC RELIGIOUS ORIGINS 103

did a decade or two ago, to a non-Semitic origin. Delitzsch,
wlio held this view in 1883 and 1886, had abandoned it as
long ago as 1889.^ That so admirable a scholar as Driver
can still hold this view is no doubt due to the fact that
the historical and religious bearings of the problem have
not sufficiently claimed his attention. It can hardly be
regarded as probable that a divinity so primitive as Ishtar
and whose cult was so widely diffused as to be wor-
shipped in every Semitic territory, should be known in
them all by a name borrowed from a foreign source by
one of the Semitic nations after the Semitic dispersion
had begun. Such an improbable view ought not to be
maintained if a Semitic etymology which is even plausible
can be suggested. That this borrowing did not occur is
indicated by the presence of the letter 'ayin at the begin-
ning of the name in all the Semitic languages except the
Babylonian- Assyrian. From this latter tongue the 'ayin
had disappeared, and it is hardly conceivable that the
'ayin would be present in all the other languages if the
name had come to them from a Babylonian source.

Most scholars regard the name as of Semitic origin and
have offered for it various etymologies. ^ The etymology
proposed in my " Ishtar Cult " derived the name from the

1 Cf. his Hebrew Language viewed in the Light of Assyrian Sesearch,
p. 11, and Prolegomena eines neues Hebraisch-Aramaisches Worterbuch,
p. 138, with his Assyrian Grammar, p. 181.

" See, e.g., Haupt, ZDMG., Vol. XXXIV, p. 758, and Moore's " Ashto-
reth " in the Encyc. Bib. in addition to references below. The following
proposed etymologies are worthy of mention: 1. Haupt holds that it is a
feminine form of the stem of the name of the Assyrian god Ashur (As-
shur), Ishtar being written for Itshar (quoted by Tiele, Actes d. 6me
Gong. Inter, d. Orient., Pt. II, p. 497, n., and reiterated by Haupt, Amer.
Jour, of Philology, Vol. VIII, p. 278, n., also his Assyrian E Vowel, 1887,
p. 16, n.). If the name of the god Ashur be derived, as is supposed below
(Chapter V), from the 'ashera or post which marked the limits of the
primitive sanctuary, this view would be plausible, were it not for the con-
fusion which it assumes from 'aleph to 'ayin. That confusion, however
natural in Assyrian and Babylonian, can hardly have occurred in primi-
tive Semitic. 2. Jensen (Zeit. f. Keilschr. For., Vol. I, p. 306), and Zim-
mem (JBabylonische Busspsalmen, p. 39), hold that the t is inserted after

104 SEMITIC ORIGINS

root 'athara, " to fall," ^ and took the name to be a reflex-
ive with both a transitive and an intransitive meaning;
the former of which meant to "cast forth" or "cause to
fall," applying to the mother ; the latter, " that which is
cast forth," or offspring, being applied in Deut. 7''' 28*' ^
to lambs. To this derivation Driver objects^ — and the
objection is a forcible one — that the root 'athara means
not simply " to fall," but " to stumble," " to trip," which,
he urges, makes the etymology unsatisfactory. This ob-
jection applies, as I now think, not to the etymology itself,
but to the particular meaning which I attached to it.
Lagarde had pointed the way to the true solution in a
short article published as long ago as 1881,^ but when the
"Ishtar Cult" was written his article escaped my atten-
tion. He connected the name, as I did, with the root
'athara, but called attention also to some important varia-
tions in the meaning of the root which Lane had exhibited
in his lexicon,* and which the connection of the goddess
with the palm tree now enables us to appreciate. While
in Arabic literature the stem ordinarily has the meaning
" fall " or " stumble," 'dthUr means " a channel that is dug
for the purpose of irrigating a palm tree such as is termed
bdal " ; 'athr, " such as is watered by rain alone " ; and
'athir, "dust," "earth," or "mud."

The idea that Ba'al was the lord of self-irrigated land
has been shown by Robertson Smith to belong to Syria.^

the second radical and that the name is to be derived from the root hashar
= 'ashar, " to unite." This etymology is far less objectionable and much
might be said in its favor. Another one vrill, however, be found to fit the
conditions more perfectly. 3. Georg Hoffmann derives it ( Ueber einige
phon. Inschriften, Gottingen, 1889, p. 22, n.), by a like method of forma-
tion from the root 'ashar, Aramaic 'athar, " to be voluptuous," a deriva^
tion too abstract in its meaning to be primitive.

1 Hebraica, Vol. X, pp. 69-71.

^ Hastings's Dictionary of the Bible, Vol. I, p. 169, n.

s"Astarte" in Nachrichten v. d. Konigl. Oesellschaft der Wissen-
schaften zu Gott., 1881, pp. 396-400.

* Arabic Lexicon, p. 1953.

^Beligion of the Semites, 2d ed., pp. 97 ff. and 109 ff. Smith held

SEMITIC RELIGIOUS ORIGINS 105

He also held that the term ha'al land was afterward bor-
rowed by the Arabs. The term ba'al for an irrigated
palm tree would in any case be late, and must have sup-
planted an earlier term. That earlier term was, I think,
connected with the root 'athara. This is borne out by
the statement which Robertson Smith cites from Bokhari
(Bulac vocalized edition) which makes 'aihari synony-
mous with "what is watered by the sky and by foun-
tains."^ Smith thought that this term was derived from
the name of the god Athtar, but from our preceding dis-
cussion it seems more probable that the derivation was
the other way, or that both came simultaneously from a
common root. It is probable, therefore, that in primitive
Semitic 'athara was connected with naturally watered
land, and that 'athtar meant, in its transitive sense, " she
who waters," or " she who makes fruitful," while in the
intransitive sense it might apply to that which was
"watered," "fertilized," or "produced," and so could
come in course of time to mean "offspring," and as in
Deut. 7", etc., " lambs." ^ This view suits the agricul-
tural and social conditions which the foregoing pages
have shown to have been prevalent in Arabia. The Be-
dawi, coming in from the arid desert to a green and fruit-
that agriculture, even to the cultivation of the date-palm, was borrowed
by the Sabsans from Syria. Since the latter has been proven above to
be incorrect, it may well be that the term bdal originated in Sabsea or
Arabia. It could not even then, however, be as old as 'dthur. See
below. Chapter IV.

1 Smith, Ibid. p. 99, n 2.

2 On this view of the etymology the meaning of the Arabic 'athara,
Heb. 'ashar, and Aram, 'athar are all explained. 'Athur means accord-
ing to Lane " a pit dug for a lion or other animal that he may fall into it
in order that he may be taken." As noted in the text another form of
the word signifies "irrigating ditch for a palm tree." The idea of falling
may have become connected with the root from the catching of wild ani-
mals in deep irrigating ditches. In time this meaning supplanted the
original one. The idea of stumbling would naturally connect itself with
the root from the idea of falling into such ditches. From the idea of irri-
gation would come the idea of fertility, whence the Hebrew, Aram., and
Syr. meanings " to be rich," " voluptuous," etc., would easily follow.

106 SEMITIC ORIGINS

ful oasis, would naturally say, " the self -irrigating or fruit-
producing goddess has her abode here." That oasis thus
became to him a garden of his god, its water and trees visi-
ble representatives of his deities. The society in which he
lived was one of great sexual freedom in which the mother
was the head of the family ; he therefore naturally thought
of his gods as a mother and son. The trees in the oasis
were palm trees; the sex relations of the trees and of
human beings were all combined in his mind in the way
already described, so that the " self-waterer " was for him
the "fruit-producer," "the creator of life," "the mother
goddess," the goddess of love. Such was the religion of
the Semites in their primitive home while they were yet
one people.

Robertson Smith's picture of their sanctuaries, their
gods, their totemistic clan organization, their sacrifices,
and their animistic conception of spirits, abides; we only
see more clearly that the chief deity of the clan was at
this primitive time a goddess, and that in so far as a male
deity played any considerable part he was her son and
reflex.^

No doubt it will be distasteful to many to believe that

II have pointed out elsewhere (article " Asherah" in Jewish Encyclo-
pedia'), that in all probability the limits of the primitive Semitic sanctu-
ary vfere marked, even before the Semitic dispersion, by wooden posts
called " asherahs " or " athrahs," and that in course of time the name of
the post was in certain localities, as in South Arabia and the Lebanon re-
gion of Syria, applied to the goddess herself. The evidence for this state-
ment is that asherah means "sanctuary" in the Phoenician inscription
from Ma'sub, and its philological equivalent ashirlu, ashrdti or eshritu,
eshr&ti is commonly used in Assyrian- for sanctuary, while a goddess Ath-
erat has been found in a Minjean inscription (Hommel in Expository
Times, Jan., 1900, p. 190), and a goddess Ashirta in the region of Leba-
non (Sayce, ZA., Vol. VI, p. 161, Epping and Strassmaier, ZA., Vol. VI,
p. 241 ; Winokler in Schrader's KB., Vol. V, p. 124, and passim ; Reisner
in Mittheilungen of the Berlin Museum, p. 92, and Jensen, ZA., Vol. XI,
p. 302H.). Cf. also below. Chapter VI. Hommel (op. c)«.) fancies that he
sees in the original form of the ideogram for Ishtar a post on which hangs
the skin of an animal (cf. Thureau Dangin's VEcriture cuneiforme, No.
294, and Smith's Religion of the Semites, 2d ed., p. 435 fl.

SEMITIC RELIGIOUS ORIGINS 107

the beginnings of Semitic religion as they were conceived
by the Semites themselves go back to sexual relations. It
must be remembered that such things were thought of
and treated much more innocently in primitive times than
would be indicated by a similar treatment now. In reality,
too, the Semite actually hit upon a feature of human life
which is, as scientific investigation is showing us, inti-
mately connected with religious feeling at the present
dayi and has had more real influence in developing
moral, altruistic, and humanitarian feeling in the past than
any other. The prolongation of the period of helplessness
in infancy and the consequent development of maternal
love out of which feelings of obligation and conscience
have grown is now seen to lie at the root of the moral and
religious progress of the race.^ The primitive Semite's
conception of his goddess and her service, to which he
attributed the beginnings of intelligence and civilization,
was in a rude, blind way an emphasis of the same truth.
Considering the animal passions of human nature, it is
little wonder that the processes of procreation often at-
tracted more attention than the offspring itself ; but the
delight which all Semites took, and still take, in their
children, is witness to the fact that such religion was
never wholly degenerate. Semitic sacrifice, commensal as
Robertson Smith has shown it to be,^ embodies in a gross
way the principle of the religious life which is expressed
in the highest spiritual form in John 17^ : " I in them
and thou in me that they may be perfected into one ; " so
the Semitic conception of deity as we have traced it em-
bodies the truth — grossly indeed, but nevertheless em-
bodies it — that " God is love."

This religion, containing a kernel of perpetual truth,

1 See Leuba in Journal of Psychology, 1896 ; Starbuck's Psychology
of Peligion, New York, 1900, Part I, on conversion ; and Coe's Spiritual
Life, New York, 1900.

2 See Drummond's Ascent of Man, New York, 1895, chs. vii and viii,
and Fiske's Through Nature to God, Boston, 1899, pp. 96-130.

' Beligion of the Semites, Lectures VII to XI.

108 SEMITIC ORIGINS

although it was formulated thus crudely, formed the sub-
stratum of the religion of the Semites in historical times.
It was modified here and there by economic changes and
the consequent change in social conditions which followed.
At other times foreign influences combined with these to
effect a transformation. In Israel its baser elements were
eliminated by the prophets, who erected on its foundation
a structure of spiritual religion.^ Traces of these primi-
tive conceptions appear throughout the Semitic world as
witnesses to the perpetual influence of these fundamental
conceptions of religion and life ; and owing to its influ-
ence through the Phoenicians upon the Greeks and through
Greek society upon the early Christians, and also its in-
fluence on the Hebrews and through them upon the
church, its effects in many ways abide to the present hour.^
In all Semitic life, religious and social, the hag or re-
ligious festival has always played an important part.
Among the ancient Hebrews there were three such festi-
vals which all readers of the Bible will readily recall, — the
Passover, near the vernal equinox, the feast of Weeks at
the end of the harvest, seven weeks after the Passover,
and the feast of Ingathering or Tabernacles at the time of
the grape harvest in the seventh month. Of these, recent
Biblical scholars regard the first only as primitive, and
hold that the others were agricultural festivals adopted by
the Israelites after the settlement in Canaan. ^ There is
much evidence, however, to show that two of these three
festivals have their roots in primitive Semitic practices, and
that what the settlement in Canaan did for them was not
to originate them, but to give them a new interpretation.

1 See below, Chapter VII.

2 See below, Chapter VIII.

' Wellhausen, Prolegomena zur Geschichte Israels, 5th ed., 1899,
p. 91 ; Reste arabische HeidenUtms, 2d ed., p. 98 ; W. Robertson Smith,
Prophets of Israel, 2d ed., pp. 38, 56, and 384, also Old Testament in the
Jewish Church, 2d ed., pp. 240, 269; Harding in Hastings's Dictionary
of the Bible, Vol. I, p. 860 ; and Budde, Religion of Israel to the Exile,
p. 73.

SEMITIC RELIGIOUS ORIGINS 109

All scholars agree that the paschal portion of the Pass-
over festival, as distinguished from the unleavened bread
features of it, existed in the nomadic life of pre-Canaan-
itish days. This sacrifice of a sheep occurred in the
month Nisan, i.e. in the spring, or at the beginning of
the Oriental summer. Similarly in Cyprus, as we learn
from Johannes Lydus,^ a sacrifice of a sheep was made
to Ashtart. This occurred also in the spring, on the 2d of
April. In Babylonia there was also a New Year's festival,
which was held in Nisan, which, at different times and in
different places, was associated with different gods. When
we can first trace it in the days of Gudea, it is the festival
of Bau,^ one of the mother goddesses, into which the
primitive Semitic mother goddess had developed in the
peculiar Babylonian conditions.^ Later, in consequence
of the forces which wrought the transformations described
below,* it appears as a feast of Marduk of Babylon.^ In
the earlier time when we can trace it as a festival of the
goddess, the offerings were lambs, sheep, cattle, etc.
Wellhausen, Robertson Smith, and Winckler have shown
that in Arabia the festival in the month Ragab originally
corresponded both in time and in character to these spring
festivals among the other Semites.® Two characteristics
are common to all these festivals, — they occurred in the
springtime, and they involved the sacrifice of lambs. In
Arabia the domestic animals bring forth once a year, and
the yeaning time is in the spring.'^ In Ex. 34, the Yah-

1 Cf. his De Mensibus, Bk. IV, 45, and Hebraica, Vol. X, p. 45.

2 See KB., Vol. IIIi, pp. 59, 61, 69, and 71 ; also Jastrow's Beligion
of Babylonia and Assyria, pp. 59 and 677 ff.

' See below, Chapter V.

* Chapter V.

^ KB., Vol. IIP, p. 15, and Jastrow, ibid., p. 677.

8 Wellhausen, Beste arabische Heidentums, 2d ed., p. 97 fl.; Robert-
son Smith, Religion of the Semites, 2d ed., p. 227 ff.; and Winckler,
Altorientalische Forschungen, 2te Reihe, Vol. II, pp. 324-350, especially
p. 344. On the character of the offerings at the Regab feast, cf. Smith,
ibid., n.

' Doughty's Arabia Deserta, Vol. I, p. 429.

110 SEMITIC ORIGINS

wistic Decalogue, the earliest of existing Hebrew law-
books, this spring festival is connected with the gift of
firstlings to Yahwe (vv. 18-20). There can be little
doubt, in view of these facts, that originally the nomadic
Semites kept a spring festival to the mother goddess of
fertility. The lambs, kids, and young camels were her
gifts, and to her it was right that a joyous feast should be
held in honor of her gracious blessings.

The circumcision festivals which were witnessed bj
Doughty occurred at the same time of the year.^ Those
feasts are still accompanied, as we noted above, by the
sacrifice of a sheep, the dancing of girls, and the selection
of wives. We cannot, therefore, be far wrong in regard-
ing them as a survival of this old spring festival. As
already pointed out, Ephraem and Augustine described
the festival of the Semitic mother goddess, as it was
known to them, as lewd.^ Originally, therefore, the
spring festival was accompanied by the sacrifice of maiden
virtue, — a sacrifice out of which grew the custom de-
scribed by Herodotus,^ as well as the sacrifice of the fore-
skin of youths. Probably acts of free love on the part
of all were also a part of the primitive ritual.*

The spring festival in this far-off primeval time was
then an occasion when the mother goddess was honored
by sacrifices to her of some of all her many gifts of ani-
mal fertility in the ways which were thought to be pleas-
ing to her. The time was appropriate, since she was
revealing in the spring her power through the offspring
of the flocks and herds, through the flowering date-palms
where her acts of fertilization were taking place, and
through the nature which she had given men.

So the infant was consecrated to her service by circum-

1 Doughty's, Arabia Deserta, Vol. I, pp. 340-342.

2 Ephraem, Opera, Vol. II, p. 458 ff., and Augustine, De CivUate Dei,
Bk. II, 4. Cf. Hebraica, Vol. X, pp. 61 and 59.

8 Bk. I, 199. Cf. Hebraica, Vol. X, p. 20.

* Such, as I take it, was the original meaning of the dance described
Toy Doughty, ibid., p. 341.

SEMITIC RELIGIOUS ORIGINS 111

cision, the maiden by the sacrifice of her chastity, and all
by acts of free love. At the same time the bonds of tribal
kinship were more closely knit by the commensal meal,
which was no doubt accompanied by boisterous manifesta-
tions of joy, and by songs, which would be extremely
coarse when judged by the more refined standards of later
ages.

Wellhausen has made it tolerably clear that in the pre-
Islamic days the Arabs divided the year roughly into
halves,^ and that the second half which originally began
in the autumn was inaugurated by the Safar festival as
the other half was by the Ragab festival. This feast he
coordinates with the Hebrew feast of Tabernacles, which
came in the month Tishri and which represented to the
Palestinian Hebrews the conclusion of the grape gather-
ing. No trace of this feast is found in Babjdonia, though
Jastrow conjectures ^ with some probability that at some
time a sacred New Year's feast occurred in the autumn
in some part of Babylonia, since the Jews, who derived
their method of reckoning time from thence, begin their
New Year with Tishri. The character of this feast among
the primitive Semites it is not hard to guess. The har-
vest of the date-palm comes at just this time,^ when the
Arabs give themselves to gladness and hospitality,* and
the nomads visit the oases to lay in a supply of dates for
the winter.^ We cannot doubt but that in ancient times
such an occasion was made a festival to the goddess of
the palm tree or that it was characterized by orgies such

^ Beste arabische Heidentums, 2d ed., p. 96 ff. Cf. Winckler, Alt-
orientalische Forschungen, 2te Reihe, "Vol. II, p. 344, who makes the
same division as V^^ellhausen, but makes it begin with Muharran, the
month before Safar.

2 Beligion of Babylonia and Assyria, p. 681. Cf. Muss-Amolt in
JBL., Vol. XI, p. 160 ff.

s See Doughty's Arabia Deserta, Vol. I, pp. 557 and 561 ; also Zwemer's
Arabia, p. 125.

* Wellsted's Travels in Arabia, Vol. II, p. 122.

^ Doughty, ibid., as n. 3.

112 SEMITIC ORIGINS

as would befit the rejoicings of a people possessing such
a social organization and pervaded by such religious ideas.
In the earliest times the oases were himas^ or tracts sacred
to the gods ; the gathering of the dates took place there-
fore in a sacred tract as well as from a sacred tree and
would accordingly be naturally regarded as a religious
act. This autumn festival still survives in Abyssinia. It
has been Christianized and is called Mascal, or the Cross.
It is celebrated in September, and a part of its ritual
includes the lighting of fires on high places before dawn,
when oxen are slaughtered as in a heathen festival. It is
celebrated, too, with dancing, drumming and playing the
sistra during the whole night. '^ Considerable elements of
heathenish rites have entered into all the phases of the
ritual of the Abyssinian church, but it is not difficult to
detect the source whence this feast has come.

Of a third festival we cannot be so confident. If it
existed in primitive times, it must have been connected
with the god Tammuz. Traces of a festival of .the god
Tammuz, preceded by wailing for him, are found in Baby-
lonia, Palestine, and PhcEnicia. It appears from the poem
known as " Ishtar's Descent," that there was in Babylonia
a " day of Tammuz." ^ It is usually held, since the fourth
of the Babylonian months bore the name of this god, that
it was then that his festival was celebrated, and Jastrow
on this basis holds that it was a solar festival, celebrated
in the fourth month at the approach of the summer sol-
stice.* He, like many others, connects this feast, which
was preceded by wailing for the death of the god and
celebrated by rejoicings at his resurrection,* as significant

1 Smith's Religion of the Semites, 2d ed., pp. 112, 142-144, and 156-
157 ; Wellhausen, Heidentums, 2d ed., p. 105 ff.

2 See Bent's Sacred City of the Ethiopians, pp. 53, 83, 84.

3 Cf. IV R., p. 31, rev. 1. 56 ; Hebraica, Vol. IX, p. 151 ; and Jeremiaa's
Zeben nach dem Tode, p. 23.

* Beligion of Babylonia and Assyria, p. 682.

' Cf . Luoian, De Syria Dea, § 6 ; Hebraica, Vol. X, p. 31 ; and
Pietschmann's Oeschichte der Phoenizier, p. 219.

SEMITIC RELIGIOUS ORIGINS 113

of the annual death of vegetation, which on Jastrow's
interpretation would be due to the burning heat of the
summer sun. It appears, however, that in Phoenicia and
Palestine the festival was celebrated not in the fourth but
in the sixth month. Ezekiel (ch. 8^) dates it according
to the Massoretic text at that time, though the LXX
place it in the fifth month. Many modern scholars follow
the LXX, but, as it seems to me, without sufficient reason.^
The cuneiform non-Semitic expression for the sixth month
was "the month of the message of Ishtar," as though it
was then that she descended to the lower world.^ The
name of the sixth month, Elul, has been explained from
the wailing for Tammuz,^ and altogether it seems prob-
able that the wailing originally occurred in the sixth
month, and was followed by the festival of date harvest
at the beginning of the seventh, of which we have already
spoken. If this be the case, the sacrifice of chastity of
which Lucian speaks in connection with these rites at
Biblos was a survival from the rites of joy with which
the date harvest was celebrated in primitive Semitic
times. That the feast of Tammuz should in some form
go back to primitive Semitic conditions is indicated by
the myth which makes Tammuz the son of Ishtar and
which, as we have noted, could only have been formed in
a society organized on the lines of the so-called matri-
archal clan. Winckler's conclusions as to the old Arabic
calendar include the opinion that there was in Arabia a
similar summer festival in July-August.^ The special
characteristics of this festival are not clearly known. It
seems likely, however, that it was a survival from the old
wailing for the death of vegetation which preceded the
glad festival of the date harvest. Primarily, then, this
feast was a sort of lent preceding the glad time of the

1 Cf. Toy's Ezeldel, in Haupt's SBOT.

» Cf. Muss-Arnolt in JBL., Vol. XI, pp. 88, 89 ; and Brunnow's Classi-
fied, List of Cuneiform Idiographs, No. 10759.

3 AUorientalische Forschungen, 2te Reihe, "Vol. II, pp. 336-344.

114 SEMITIC ORIGINS

autumn festival, when the tree of Tammuz and Ishtar
yielded its fruit.

Following Robertson Sniith,i I expressed in the "Ishtar
Cult" the opinion that the wailing for Tammuz was ori-
ginally the wailing for a sacrificial victim.^ I still incline
to think that this view is right, although, as then, I
think that at a very early period it may have received a
new explanation which connected it with the death of
vegetation. In the deserts of Arabia when the burning
summer sun dries up the pastures and in consequence
the milk of the domestic animals largely fails, while the
summer heat renders life almost unendurable,^ it may
well have seemed to the nomads that Tammuz was dead.
Thus the wailing, which originally accompanied the death
of the victim at the festival, was, I think, extended to
cover a portion of time preceding harvest. This produced
a period of gloom to be turned to life when harvest came
with its evidences of the god's returning life.

Robertson Smith has with great plausibility connected
the fasting and humiliation of the Jewish Day of Atone-
ment with this Tammuz wailing.* Such connection is
from every point of view exceedingly probable. The Day
of Atonement came at the beginning of autumn, a fact
which confirms our view that it originally occurred in
connection with the autumn feast.^

If this view be correct, it is not difficult to understand
how the Tammuz wailing and ritual may have been trans-
ferred in Babylonia to the fourth month. The first har-
vest of wheat and barley is in that country reaped at the
time of the summer solstice, and at such a time a festival

' Beligion of the Semites, 1st ed., p. 392, n.

2 Hebraica, Vol. X, p. 74.

8 Doughty's Arabia Deserta, Vol. I, chs. xvii and xviii, esp. p. 472 ff.

* Seligion of the Semites, Leot. XI, esp. p. 411.

5 Fraser, Golden Bough, ch. iii, connects the death of Tammuz with
the corn (wheat) harvest, — the slaying of the divine grain. This cannot
have been primitive, on account of the economic conditions of Arabia,
though possibly it was a later agricultural explanation.

SEMITIC RELIGIOUS ORIGINS 115

among an agricultural people is a most natural occurrence.
If in order to meet this need the Tammuz festival were
put forward a few weeks, the influence of Babylonia on
Palestine in the El-Amarna period would lead (if local
influences had not already done so) to the establishment
of a festival at the end of harvest there. This afterwards
the Hebrews adopted as the feast of weeks. Meantime
the direct influence of Arabia seems to have been sufficient
in Phoenicia and Palestine to keep the original Tammuz
festival at its own period in the autumn separate from the
festival at the end of barley harvest. Something like this
may have been the course of development in Babylonia.
The fact that the fourth month bore the name of Tammuz
is a somewhat slight basis for such conjecture, since the
month may have been given the name for other reasons.

We conclude, however, that but two Semitic festivals
were primitive, the festival of the yeaning time in the
spring and the festival of the date harvest in the autumn.
Out of these the other festivals of the Semitic world have
been developed, except as some of them have been bor-
rowed from the peoples of the lands in which they settled.

If now we turn to the Hamites, from whom originally
the Semites separated themselves, we find some indications
that their primitive institutions were similar to those of the
primitive Semites, if not identical with them. Circum-
cision was, as we have already noticed,^ practised by the
Egyptians and the Hamitic Gallas, and Nowack^ and Ben-
zinger ^ still hold with Herodotus that the Semitic rite was
borrowed from Egypt. Down to the time of the Csesars
women and girls were licensed to a life of immorality by
consecration to the service of Amon at Thebes. These
women were held in such high esteem that this public
course of life did not prevent them from making good
marriages when age compelled them to withdraw from
this service.* Maspero interprets this rightly as a relic of

1 See above, p. 101. ' Archceolorjie, p. 154.

2 Archceologie, Vol. I, p. 167. * Strabo, Bk. XVII, 46.

116 SEMITIC ORIGINS

polyandry. 1 In its later stages this polyandry was endog-
amous, like the later polyandry of the Semites, since it
permitted the marriage of brother and sister, and some-
times of father and daughter. One of the legacies left to
Egypt by this type of polyandry is the use of the words
"brother" and "sister" in the sense of "lover" and "mis-
tress."^ This stage of the civilization is further indicated
by the fact that in the temples of the chief gods there
were women devoted to purposes similar to those for which
they were attached to the temple of Amon, while in the
temples of the female divinities they held the chief
places.^

Another reflection in the Egyptian religion of this state
of society is found in the conception of the goddess Isis.
The oldest myths concerning her represent her as an inde-
pendent deity, dwelling in the midst of the ponds without
husband or lover, who gave birth spontaneously to a son,
whom she suckled among the reeds,* — a tale which can
only be properly interpreted as the reflection of a society
in which exogamous polyandry, of a type resembling the
earliest Semitic polyandry, was practised. In later times
she was said to be married to Osiris, a fact which one
school of mythologists interpret as the marriage of Isis,
the Dawn, to Osiris, the Sun,^ but which Maspero, with
more probability, takes to mean the marriage of Isis, the
Earth, to Osiris, the Nile.® There can be no doubt but
that this latter myth is the later of the two, and is the
reflection of a later social organization.

We have then in the oldest Hamitic civilization traces
of circumcision, of polyandry, of a mother goddess who
represented well-watered land, as among the Semites.

1 Masp^ro's Damn of Civilization, p. 50.

2 Maspero, op. cit., p. 51.

» Maspero, op. eit. , p. 126.
' Maspero, op. cit., p. 131.

5 So Robert Brown, Jr., Semitic Influence in Hellenic Mythology, p. 63.
Brown is a disciple of Max Mttller in the interpretation of mythology.
« Masp6ro, op. cit., p. 132.

SEMITIC RELIGIOUS ORIGINS 117

The date-palm was also known, and there is at least one
trace of it as a god.^

Among the Hamites who lived to the west of Egypt
similar customs appear. Thus Herodotus tells us^ that
the Nasamones, a tribe of Berber Hamites,^ made yearly-
expeditions to a date-palm oasis to gather the fruit, and
their polyandry and sexual customs in general resembled
closely much which we find among the Semites.

With reference to these institutions which the ancient
Semites and Hamites had in common, there are three
possible opinions : (1) they may have developed them
in the early days before the two peoples separated, when
as yet the races were one ; (2) they may have developed
them independently through the influence of similar en-
vironments ; or (3) one race may have borrowed from the
other at a comparatively late period.

The last of these possibilities must be rejected at once.
We have shown above how all these institutions of prim-
itive Semitic life, including even circumcision, grew nat-
urally out of the desert and oasis life such as they were
subjected to in Arabia. It is purely arbitrary, therefore,
to assume without positive proof that any one of these
institutions was a late intruder into Semitic practice. The
theory of Herodotus with reference to circumcision must
therefore be abandoned. On the other hand, few will be
found to maintain that it or any of the other institutions
under discussion were borrowed by the Egyptians from
the Semites. A people which reached such a high state
of culture at such an early epoch is not likely to have
borrowed a religious and social practice from so rude a
people as the Semitic Arabs at a time when the two must
have been separated by sea and desert.

Of the other two possibilities, the first is, under the
circumstances, by far the most probable. While, of course,

1 Masp^ro, iUd., pp. 27 and 121, n. 1.

2 Book IV, 172.

» Sergi, Mediterranean Race, p. 47.

118 SEMITIC ORIGINS

two peoples of kindred race may in similar environments
have developed similar institutions independently of one
another, it must be remembered that the environment of
the Egyptians, from the time of their settlement in Egypt,
was not similar to that of the Semites, or of a character to
produce institutions similar to theirs. Egypt is not a land
of oases, but a river-land similar to Mesopotamia. It was
an agricultural country, rich and productive. As we shaU
show below, the civilization produced in such a land was
not polyandrous, and differed consequently, as to all the
features which grow out of polyandry, from that which
the desert-oasis life produced. North Africa, outside of
Egypt, was for the most part a barren country, with occa-
sional oases, in its general features not unlike Arabia.^ It
is altogether probable that, as these regions filled up, con-
ditions were produced by the crowded populations similar
to those which we have proven for Arabia,^ and that
in consequence a similar culture of the date-palm,^ a
similar organization of the clan, a similar worship for the
feminine productive principle, and in general, similar in-
stitutions were in some portions produced, though the
fertile valleys in some portions of North Africa probably

1 Por a description of North Africa and its oases, see The International
Geography, ed. by Hugh Robert Mill, London, 1899 ; for Morocco, p. 905 ;
for Algeria, p. 907 ; for Tunesia, pp. 913, 914 ; for Tripoli, p. 916 ff.

2 Above, p. 73 ft.

3 My friend, Professor W. Max Miiller, tells me that the whole Paradise
story of Genesis, which, as we have seen, reflects primitive Semitic ideas,
has a parallel in the hieroglyphic Egyptian. This indicates that what we .
have proven for the primitive Semitic conceptions of religion which grew
out of oasis life, could in like manner be proven for the Hamites. In
other words, the institutions which we have proven in Arabia were born
earlier in North Africa. Sayce (PSBA., Vol. XXII, p. 278) describes
a vase taken from a predynastic tomb, on which a palm tree is pictured.
It is, therefore, unnecessary to account for such phenomena in the Egyp-
tian religion on the ground of Semitic influence from Arabia. It is in the
highest degree doubtful whether such foreign influence, exercised apart
from conquest or settlement, produced such results anywhere in the
ancient world. Semitic words in later Egyptian inscriptions are no argu-
ment against this view.

SEMITIC RELIGIOUS ORIGINS 119

prevented the production of these institutions on so wide
and so uniform a scale as in Arabia.

Now such crowding of the country must have occurred
before the Semitic migration, and must have been its cause.
Some such force must have impelled the first immigrants
to enter the unattractive Arabian peninsula. We have,
then, in the primitive Hamito-Semitic home the elements
present for the birth of these institutions before the separa-
tion of the two grand divisions of the race. We hold it
probable, therefore, that the totemistic clan, the culture of
the date-palm with its worship, the mother goddess as the
typical divinity, and circumcision, had to some extent their
beginnings at the time when the Hamites and Semites
were living in that common home of their infancy, in
which their kindred tongues were born, notwithstanding
that the differences in those tongues bear witness to the
fact that they separated in prehistoric time, thousands of
years ago.

If this be so, we find, by applying Hilderbrand's law
referred to above,^ that, at the time of the separation of
the Semites from the Hamites, the pastoral and semi-
agricultural stage of life had been reached, with some rude
cultivation of the date-palm. This conclusion removes
the time of Hamito-Semitic savagery some thousands of
years farther back into remote antiquity than it has usually
been placed, but it is not on that account to be rejected.
It must, it seems to me, be regarded as highly probable .^

The conclusions reached in this discussion inevitably
lead to another, of no little importance to the correct un-
derstanding of the Semitic religions. If we regard it as a

1 pp. 72, 73.

2 The writer makes no pretence to a knowledge of Egyptology. The
facts quoted above are quoted on the authority of reputable Egyptologists.
An interpretation has been given them in the text, such as is compatible
with the results made probable by the interpretation of the Semitic
material in the light of economic and social laws. It is to be hoped that
some Egyptologist will take the matter up and do the same for the
material of his science.

120 SEMITIC ORIGINS

law, that religious institutions are in some important
respects patterned on those social and political institutions
which economic environment makes possible, we should
naturally expect the Semites, as they modified their en-
vironment or moved to new ones and developed a system
of male kinship, to make masculine instead of feminine
deities the chief objects of their worship.

The feminine deities thus displaced were the earliest
principal deities which the Semites had, for even in their
savage state, their monogamy was too temporary to permit
of a system of male kinship. We regard it, therefore, as
a general principle which may be safely applied, that those
phases of Semitic religion which reflect a polyandrous
state of society are more primitive than any of those
which reflect a patriarchal. The latter are either of later
birth, are borrowed from foreign peoples, or are formed
from a mother goddess, by changing her gender but retain-
ing many of her attributes. As society, in consequence
of changed environment, was transformed from the matri-
archal to the patriarchal form, such transformations of
deities actually occurred, as will be shown in the next two
chapters. Here and there the old mother goddess survived
in something of her pristine independence, preserved by
the forces of religious conservatism. Where transforma-
tions into a masculine deity occurred she was often in part
retained as the consort and companion of the male deity.

Thus by classifying the deities on this principle and
following out the lines of economical and social develop-
ment, much light will be thrown upon the problems of
Semitic religion, and gods which have been considered to
be borrowed by one Semitic nation from another will
frequently appear to be independent developments from
a common mother goddess of the primitive time.^

1 It does not fall within the scope of this investigation to account for
the origin of the idea of a god or of supernatural spirits among the primi-
tive Semites. It is probable, however, that among them religion did not
originate in ancestor worship. Cf. Frey's Tod Seelenglaiibe und Seelen-

SEMITIC RELIGIOUS ORIGINS 121

Tcult in alien Israel, and Gruneisen's Der Ahnencultus und die Urreligion
Israels. For arguments on the other side, cf. Stade Geschichte, Vol. I,
pp. 387-427 ; Schwally, Leben nach dem Tode, ch. I ; Charles, Escha-
tology, p. 20 ff., and G. A. Smith, Preaching of the Old Testament,
p. 184, n. 2.

CHRONOLOGICAL TABLE

South Semitic Chronology
Minsean kingdom, cir. 1250-600

B.C.

Sataean kingdom, cir. 750-116 b.c.
Kingdom of Saba and Eaidan, 115

B.C. to cir. 350 a.d. (south Arabia

and Abyssinia).
Kingdom of Aksum, cir. 350 a.d.

onward (Abyssinia).
Mohammed born, 571 a.d.
Mohammed began preaching, 610

A.D.

Mohammed left Mecca, 622 a.d.
Mohammed's death, 632 a.d.
Medina Caliphate, 632-661 a.d.

West Semitic Cheonologt

El-Amama Tablets, cir. 1400 b.c
(Tyre, Sidon, Gebal, Jerusalem,
Laoish, and Ashtoreth then flour-
ished).

Kingdom of Tyre, cir. 1300-332 b.c.
(Hiram, king, cir. 1000; Eth-
Baal, cir. 880 ; Baali, cir. 660).

Israel invaded Canaan, cir. 1200

B.C.

Yakhwemelek, king of Gebal, cir.
400 B.C.

Eshmuuazer, Tabnith, and Esh-
munazer II, kings of Sidon, cir.
380-332 B.C.

Greek rule of Phoenicia, 832 b.c
onward.

Phoenician kings of Citium and
Idalion in Cyprus, 479-312 b.c

Grseco-Egyptian kings rule in Cy-
prus, 312 B.C. onward.

Carthage founded, cir. 825 b.c

Carthage subject to Rome, 201 b.c

122

CHAPTER IV

TRANSPOKMATIONS AMONG THE S0T7THEEN AND
WESTERN SEMITES

We have traced above ^ the general lines of social devel-
opment in Arabia and have noted how the vague polyandry
of the Nair type and descent through the mother were,
through economic causes, and possibly the influence of war
and marriage by capture, transformed into haal marriage
and descent through the father ; polyandry of the Thibetan
type forming in certain localities an intermediate stage of
social development. Robertson Smith is no doubt right
in holding that this transformation did not take place
before the Semitic dispersion. ^ Much evidence will be
presented in Chapter VI in support of this view. The
transformation had, however, taken place by the time of
Mohammed, who provided that wives whom their husbands
could not trust might be rebuked, secluded in lonely apart-
ments, and even flogged by the husband.^ Of course, in
parts of Arabia the older liberties of women may have
been retained much longer than they were at Mecca and
Medina, but the trend of social development, as the pas-
sage quoted above* from Strabo indicates, had been for
some time in the direction of the patriarchal family. It
must have been well advanced in Mecca and Medina
before the prophet could make such a law as that
referred to.

This transformation of the family and the exaltation of
the father left its impress upon the Arabic conception of

1 Chapter II. 2 Kinship, p. 179. s Qur'an, 43".

* p. 64. See Strabo, Bk. XVI, p. 783.
123

124 SEMITIC ORIGINS

divinity. In South Arabia, the old mother goddess, even
while she retained her feminine name, became a mas-
culine deity and a father god.^ He is frequently desig-
nated as lord (ba'al) Athtar, and is at least once called
"father." 2

In Yemen, in the southvsrestern corner of Arabia, this
impress can be most clearly traced. The country is of
volcanic formation, consisting of extensive uplands, broken
by mountain ranges and interspersed with valleys of sur-
passing richness, where from time immemorial the land
has been laid out in terraces, the water of the rainy season
stored in cisterns for irrigation, and many natural rivulets
course down the hills. ^ These valleys produce wheat,
barley, maize, millet, and coffee, as well as palm trees,
orange, lemon, quince, mango, plum, apricot, peach, apple,
pomegranate, and fig trees. The vine also grows there
luxuriantly.* This is the Arabia Felix of the ancients.
Here a Semitic kingdom had been established, probably as
early as 1250 years before the Christian era, and perhaps
earlier. The claims of Glaser and Hommel that a Mingean
kingdom preceded the Sabsean on this soil seem to me
to be well made out. The Minaean sarcophagus of the
Ptolemaic period, discovered some years since in Egypt,^
is no objection to this ; it only shows that the city of Ma'in
kept its identity some time after it was dominated by the
Sabaean power. Tradition has it that a queen of Sabsea
visited Solomon,^ and Sargon, king of Assyria, counted
It'amara (Jetha'-amara), king of Sabsea, among his tribute-

1 Cf. Hebraica, Vol. X, pp. 52-59 and 202-205.

^Mordtmann's ^imjanscfte Inschriften und Alterhumer-Mittheihingen
kg!. Museen zu Berlin, Heft VII, No. 862.

" Cf. Reolus, JTie Earth and its Inhabitants, New York, 1885, Vol. IV,
p. 438 ff., and Zwemer, Arabia, the Cradle of Islam, chs. v and vi.

* Zwemer, op. cit., p. 57.

6 Cf. Golenlscheff, in the St. Petersburg Sapiski, 189.3, p. 219 ff. ; T). H.
Miiller, in Wiener's Zeitschrift f. d. Kunde des Morganldndes,\?,M, p. 1 ff. ;
Hommel, F8BA.,Yo\. XVI, 145 ff. ; Derenbourg, in Jour, asiatique, 1894 ;
and Weber, Mitteilungen vorderasiat. Gesellsehaft, 1901, Heft I, p. 42.

6 1 Kgs. 101 ff.

TRANSFORMATIONS 125

payers in the year 715 B.c.,^ but it seems probable that
these references are to a North Arabian precursor of the
Sabsean kingdom.* In the rich valleys of southwestern
Arabia, agricultural communities must have been formed
at a very early time. Semitic social life "would therefore
be transformed here far sooner than in other parts of
Arabia. Reclus declares ^ that in this mountainous region
the very soil and climate render a nomadic life almost im-
possible. There are vast uplands between the mountains
and valleys where the Bedawi have settled into a pastoral
life.* To this region the Semites from central Arabia
came, here their social structure underwent in course of
time, and in consequence of their new conditions, a trans-
formation. Descent was reckoned through the father, and
in time the old mother goddess was transformed into a
god. He became, of course, a father, and is frequently
called lord (ha'al) Athtar.

Not only is this true, but we are able in one inscription
to catch a glimpse of the deity in the very process of trans-
formation. This interesting document was published some
years since by the Derenbourgs,^ and is of sufficient im-
portance to be reproduced here. It reads : —

1. Yasbakh of Eiyam, son of Mauqis and Baus, and his wife Karibat

of M. . . .

2. of the tribe Sirwakh, a royal vassal, — they have consecrated to

their lady 'Umm-Athtar for

3. four sons four images of pure gold, because 'Umm-Athtar blessed

4. them with the boys and three daughters and they lived — all these

chil-
6. dren — and they two themselves have acquired gain through these

children. May 'Dmm-
6. Athtar continue to bless his servants, Yasbakh and Karibat with

well-formed children, and to favor them

1 Cf . KB. , Vol. II, pp. 54, 85 ; and Glaser, Die Abessinier in Arabia
und Afrika, p. 29.

2 Cf. Weber, in Milteil. der vorderasiat. Gesell, 1901, I, 32.
» Op. cit., p. 438.

* Zwemer, op. cit., p. 68.

6 Journal asiatique, 8 Ser., Vol. II, pp. 256-266.

126 SEMITIC ORIGINS

7. themselves, and to favor their children. May 'Umm-Athtar be

gracious

8. and grant complete safety to the sons of Yasbakh, Kharif Mag-

da'al, Ra-

9. balat and 'Am'atiq, the descendants of Mauqis, and to their har-

vests and good fruits in
10. the land Nakhla Kharif, and in the pastures of their camels. To
'Umm-Athtar.

The value of this inscription to our subject can scarcely
be overrated. 'Umm-Athtar is not a new female divinity
as the Derenbourgs thought,^ but is as Mordtmann has
rightly seen, simply "mother Athtar."^ 'Umm-Athtar is
also in 1. 2 called "lady."^

It is therefore clear that she is, in the thought of these
worshippers, still a goddess. No doubt, therefore, it was
an Athtar thus conceived who was invoked in such proper
names as Yasma'um, meaning "may mother hear."*
Although the parents, who caused the inscription under
discussion to be written, addressed Athtar as " lady " and
as "mother," they nevertheless describe themselves as
" his servants," ^ showing that they were conscious that at
times, or possibly that in neighboring places, the deity
was regarded as a god, and that the transition was be-
ginning to make itself felt in their own thought. At the
same time the character of the old mother goddess, or
deity of sexual love and fertility, was still strictly main-
tained ; they give thanks for the birth of their seven chil-
dren, and while they pray for more they pray also for
their harvests and pasture lands, over which the same
goddess has power. It is also an interesting fact that the
name of the place where this "mother Athtar," who is
called "he," was worshipped was Nakhla Kharif, which

^ Op. cit., p. 259.
" Simjarische Inschriften, p. 25.
8 Sabsean, nxils.

4 Sabsean, D»raD\ Cf. Journal asiatique, 6 Ser., Vol. XIX, p. 213,
No. 4171.

6 Sabsean, W-\2V.

TRANSFORMATIONS 127

means " the palm tree of ripe fruit," — a fact which con-
nects this deity with the date-palm, as we have done in
the preceding chapter.

Tlie evidence that Athtar in Sabsea was transformed
into a god is abundant. He is called hdal or " lord " in a
considerable number of inscriptions which come from sev-
eral different localities,^ and is at least once, as already-
noted, called "father." Robertson Smith was of the
opinion that the term hdal originated in Syria and was
borrowed by the Yemenites from thence.^ If this be true,
the application of the term ha'al to Athtar would be
decidedly late. Smith's argument is, however, based on
the supposition that all agricultural processes were bor-
rowed from Syria, even to the cultivation of the date-palm,
— an opinion which our investigation has proven <* to be,
in part at least, untenable. The primitive Semitic social
and religious institutions presuppose the culture of the
date-palm and a semi-agricultural life. The course of
development which Semitic social life underwent, how-
ever, assures us that the bdal form of marriage made its
appearance at a comparatively late time, and that therefore
the ha'al conception of deity is likewise late whether it
was borrowed by the Yemenites from Syria or not. We
are safe, then, in assuming that the ha'al Athtar is later
than the ^umm Athtar and was developed out of her.

This masculine Athtar was in places called " lord of the
water supply," * and like the feminine Athtar described
above, was a god of fertility, whose blessing was necessary
to abundant harvests.^ That Athtar became localized in
different places in each of which slightly different con-
ceptions of him were entertained, so that there were

1 See CIS., Pt. rV, "Vol. I, Nos. 40^, 41', 466 . Mordtmann, op. cit., No.
8862, and ZDMG., Vol. XXXVII, pp. 4 and 326.

2 Beligion of the Semites, 2d ed., p. 107 ff.
8 See p. 75, n. 4.

* Cf. CIS., Pt. IV, Vol. I, No. 41, and Fell, ZDMG., Vol. LIV, p. 245.
6 CIS., Pt. IV, Vol. I, Nos. 104 and 105.

128 SEMITIC ORIGINS

different gods of this name just as there were different
Ishtars in Assyria and different Virgin Marys in modern
Europe, I have shown elsewhere. ^

It frequently happens in such cases that some favorite
epithet of the deity is used so constantly to designate him
that it finally displaces his original name: thus Tammuz
(or whatever the primitive name was) became Adon, as
did Yah we in Israel. Fell has shown ^ how frequently the
Sabseans attached epithets to their gods, but his investi-
gation of the subject is not sufficiently thoroughgoing.
He stops at what are still epithets only, and does not at-
tempt to distinguish those divine names which originated
as epithets. It is possible to show how several divine
names in Arabia originated in this way. At 'Amran the
epithet Ilmaqqahu, "the divine protector," very nearly
displaced the older name of Athtar.^ In thirty inscrip-
tions, Ilmaqqahu has displaced the name Athtar except in
two instances,* and in the former of these ^ the meaning
of the inscription equates Ilmaqqahu with Athtar. Ilmaq-
qahu is, moreover, throughout this group of inscriptions, a
protector of children and a giver of fertility, — functions
not only performed by Athtar elsewhere, as we have seen,
but also performed by Athtar in this very town, as one of
the inscriptions from which the epithet Ilmaqqahu is
omitted proves.^ This epithet was known elsewhere, e.g.
at San'a,'^ but at 'Amran it nearly displaced every other
name of Athtar. At times, as in some of the inscriptions
published by Mordtmann,® the personification of the epi-
thet goes so far that Athtar and Ilmaqqahu are put side
by side as separate gods. Thus the evolution of a new
deity from an old, by the use of an epithet, was completed.

From this phenomenon it is safe to infer that if other
South Arabian, or indeed Semitic, gods appear, whose

1 Hebraica, Vol. X, pp. 57, 203, 204. * Nos. 74 and 102.

« ZDMG., Vol. LIV, p. 238 ff. « No. 74.

8 See CIS., Pt. IV, Vol. I, Nos. 72-102. « No. 102.

' Cf. CIS., Pt. IV, Vol. I, No. 18. 8 ZDMG., Vol. LII, pp. 394-400.

TRANSFORMATIONS 129

names are epithets, and whose characteristics and functions
are clearly those of Athtar, that they are offshoots from
him and have arisen in a similar manner. Another
instance of this may be seen in the god Talab Riyam,
" The Strong One of Riyam," i (analogous to the " Mighty
One of Jacob " ^ as a name for Yah we). The S6uth Arabic
epithet has, however, gone so far that the combination of
the name of the locality with the adjective is complete, and
the two have so fully displaced the name of Athtar that
in one instance^ Athtar is enumerated in the same sen-
tence as a separate deity. That the two were originally
one appears from the fact that they both have the same
functions of fertility.*

Thus Yemen developed many masculine deities, — how
many, we do not yet know, and probably shall not when
all the inscriptions which South Arabia can yield are found
and read, for many of them no doubt passed away without
leaving any monument behind to commemorate them.

The old mother goddess was not, however, lost in con-
sequence of this transformation. She was retained as the
consort of the male deity ; or to speak more accurately,
she was divided into two deities, a masculine and a fem-
inine. This masculine deity was identified in the age from
which our inscriptions come with the morning star, and
was often known as " Athtar Sharqan," ^ while the fem-
inine deity was identified with the sun, and called " Shams."
That Shams should be a goddess in South Arabia while
Shamash was elsewhere a god, was due, perhaps, as we
shall see, to the absence of foreign influences ; but whatever
its explanation, it is a fact. The expression " to the god-
dess Shams, the Lady " (ha'alaf) occurs in one inscription,^

1 Mordtmann, Simjarische Inschrifien, etc., Nos. 825, 826, 830, 860,
866, 875, and 879.

" Isa. 49'^ and 60i6. ^ uo. 866. * Cf. No. 82527-29.

5 Cf. Hehraica, Vol. X, p. 204. Fell, however, thinks that Sehsean
analogies would lead us to regard Sharqan as a place. See ZDMQ., Vol.
LIV, pp. 241, 242.

6 Mordtmann and MuUer's Sabaische Denkmdlern, No. 13.

130 SEMITIC ORIGINS

while the term hdalat is applied to her in others.^ She is
therefore clearly feminine. She was further thought to
be the spouse of the masculine god of fertility, and they
together were thought to be the parents of their worship-
pers. Thus one inscription, published by Mordtmann,
read in its original form, as he has pointed out, " with Talab
Riyam and with Shams, their parents, are the sons of
Dinn."^ Here Shams is undoubtedly the spouse of a deity
which, as we have seen, sprung from Athtar. This pas-
sage makes it clear that when "Athtar Sharqan and
Shams" are coupled together in another inscription, as
objects of devotion,^ they were worshipped as the union
of male and female in the divine circle, analogous to the
union of husband and wife in the patriarchal home, which
had become the basis of Arabian society.

The material upon which these observations are founded
comes from the Sabaean and later periods. The same was
probably also true of the religion of the earlier Minsean
kingdom. At least that is the conclusion forced upon me
by an examination of the material accessible to me. The
Minsean kingdom was composed of a number of tribes,
each of which had its local deity. In what is probably one
of the earliest Mineean inscriptions,* coming from Waqhail,
the second king of Ma'in whose name is recovered,^ three
deities are mentioned, — Athtar of Qabd, Wadd, and Nak-
rakh. That this Athtar was not originally a member of the
Pantheon of the city of Ma'in seems certain, from the

1 E.g. in Mordtmann's Simjarische Inschriften, No. 880.

2 Op. cit., No. 869.

" CIS., Pt. IV, Vol. I, No. 74. There are two or three interesting pas-
sages in which Shams is associated with a masculine form of Athtar (in
two oases it is Ilmaqqahu) and is called "his Shams." See C/jS., Noa.
106; 143 and 149. Winckler has endeavored to show (ZDMG., Vol.
lilV, pp. 408-420) that Shams also meant "goddess" at times, as
Ishtar did among the Assyrians. Even if this be so, it, like Ishtar, also
designated a definite goddess.

* Hal^vy, No. 255 {Journal asiatique, 1872).

'" Cf. Weber, in Mitteihmgen der vorderasiat. G-esell., 1901, Heft I,
p. .59, and Mordtmann, ZDMG., Vol. XLVII, pp. 395-417.

TRANSFORMATIONS 131

fact that she (or he) is connected with another place, and
also from the fact that a king who reigned three or four
steps down the list mentions only Wadd and Muradawahi
as the deities of Ma'in.^ In the inscriptions of other kings
a number of Athtars of other places are included ; it is
natural to conclude, therefore, that at Ma'in the native
deities were Wadd and a feminine deity called by various
epithets. That the primitive Semitic order of society and
of thought had largely passed away is shown by the
presence of the la'al idea in these inscriptions.^

The name Wadd is an epithet formed from a root mean-
ing " love," and there can be no doubt that Wadd is but
another name for a masculinized Athtar, or for an Arabian
Tammuz. Another Minsean inscription makes him the
consort of the old mother goddess under the name of
Athirat,^ a name derived from the posts which marked the
old Semitic sanctuary. There can be little doubt, there-
fore, that the transformation which we trace elsewhere in
later inscriptions, had taken place at Ma'in by the begin-
ning of the period of the Minsean kingdom.

If now we pass to North Arabia and assume that the
same laws of religious development were at work there, it
will appear that a number of deities were evolved by the
same social forces out of the transformed mother goddess,
and that Allah himself, the one true God of Mohammed-
anism, was originally one of these gods. That Mohammed
introduced a large spiritual element into the Islamic con-
ception of Allah cannot be denied. The strong assertion
of the unity of God, his eternity and aloneness, made, for
example, in Sura 112, distinctly and immeasurably exalted
the Arabian conception of divinity. Mohammed's earnest
effort to make the association of any other divinity with
Allah impossible * dealt a death blow to the old heathenism.

1 Halgvy, No. 229.

2 Cf. the proper name "Ba'alat," in Hal^vy, No. 234.

' Cf. Hommel, Expository Times, Vol. XI, p. 190, and Aufsatze
und Abhandlungen, Vol. II, p. 206 ff. ; also below, Chapter VI.
* See, e.g., Suras 4 and 53, passim.

132 SEMITIC ORIGINS

No one who reads the Qur'an can doubt that for much of
the purity and loftiness of this monotheism Mohammed
was indebted to Judaism and Christianity. ^ It is also
clear that the pure monotheism of the early years of his
ministry was not attractive to his fellow countrymen, and
that in the later years of his career several concessions
were made to the older Arabian religious ideas. Thus
after his migration to Medina the qibla, or direction of the
face in prayer, was changed from Jerusalem to Mecca.^
In consequence, the Qa'aba became for the Mohammedan
the sanctuary of Allah, so that Mohammed could subse-
quently call it " the holj'' house," " the holy sanctuary. " ^
He also provided that if, while on a pilgrimage to Mecca,
one killed game, — an act which violated an old taboo,* —
a sacrifice of atonement for it must be offered to Allah in
the Qa'aba.^ The one God, Allah, was thus identified
with the god of the Qa'aba, and it became his sanctuary,
which it has remained to the present time. One of the
evidences of borrowing is that, though Mohammedanism
knows no necessity of atonement, yet a part of the ritual
to which every pilgrim to Mecca must conform is the
offering of a sacrifice.^ This is a camel, bullock, goat, or
sheep, according to the wealth of the pilgrim. Such a
custom is clearly a survival from heathen ritual.

This identification of Allah with the god of the Qa'aba
could, however, not have been made had not a god been pre-
viouslj' worshipped at the Qa'aba who could be thus fused
with Mohammed's Allah without doing serious violence to
religious feeling. It has been frequently pointed out that

1 Cf . Geiger's Was hat Mohammed aus dem Judenthum attfgenommen 9
1833, and The Bible and Islam, by Henry Preserved Smith, New York,
1897.

2 Sura 2i4».

3 Sura 52- 3.

* Robertson Smith's Seligion of the Semites, 2d ed., p. 112 ff., and
144 ff.

5 Sura S^s-ss.

Cf. Zwemer's Arabia, the Cradle of Islam, 1900, p. 39.

TRANSFORMATIONS 133

the goddess Al-Uzza was especially connected with the
sanctuary at Mecca. ^ It is clear, however, from one of
Mohammed's Meccan Suras which dates from the earlier
years of his ministry (53^®"), that the popular mythology
made Al-Uzza, Al-Lat, and Manat daughters of a male
deity which, even at this early period, Mohammed identi-
fied with Allah. This is not surprising, for what is more
natural than that Mohammed should believe that the god
of his childhood's tribal faith was after all one with the
God of his larger thought and prophetic ministry? That
a goddess should be at Mecca the daughter of a god is the
reverse of the conception which prevailed among the primi-
tive Semites, and which was preserved among the Naba-
thaeans, where Al-Lat was regarded as the mother of
Dhu-'l-Shara.^ These facts and the course of development
in the conception of deity which we have traced in south
Arabia lead us to suggest the following as the probable
history of the conception of deity at Mecca. The shrine
with its sacred spring, the Zemzem, was originally the
shrine of the mother goddess, Athtar. By processes iden-
tical with those which operated in Yemen she was divided
into a masculine and a feminine deity, or by influences
similar to those which appear in the Gilgamish epic in
Babylonia, Tammuz had become her husband. This is not
a mere conjecture, since the memory of this masculine and
feminine pair under the names Isaf and Naila has actually
been preserved in Mohammedan tradition.^ To this femi-
nine deity different epithets were applied. Whether these
epithets grew out of the thought of the Meccans them-
selves, or were in part the result of syncretism with other
Arabic tribes, we cannot now determine. These epithets
became so fixed that in time they were regarded as the

' Cf. Robertson Smith's Kinship, p. 294, Wellhausen's Heidentum
2d ed., p. 36 ft., and Hehraica, Vol. X, p. 64 ft.

2 See Smith, Kinship, p. 292 ; and Religion of the Semites, p. 56 ft. ; and
Hehraica, Vol. X, p. 64 ; and below, Chapter VI.

«Cf. Wellhausen, Heidentum, 2d ed., p. 77.

134 SEMITIC ORIGINS

names of different goddesses. In course of time the male
phase of the Athtar at Mecca so overshadowed the female,
just as the patriarchate had in human society overshadowed
the matriarchate, that the feminine deity, the mother, was
to such a degree subordinated that the male could be
called Al-lahu, or "the god." This god was also known
as Hubal, and there was an idol of him in the Qa'aba.
Wellhausen has anticipated me in identifying him with
Allah. ^ For a long time one of the epithets applied to the
old mother goddess was thought to be the name of Hubal's
spouse, while others were thought to be the names of their
daughters. By the time of the prophet, however, if we
may judge by the one allusion made to it, Allah and his
daughters were thought by the Meccans to constitute their
pantheon.

At this juncture Mohammed appeared and endeavored
to exalt and purify the conception of Allah possessed by
his countrymen. Finding himself after years of preaching
unable to banish the old heathenism, he compromised, —
banishing the goddesses who still were the patrons of social
impurity, and persuading the people to regard them as
mere names, — he led them to identify the god of the
Qa'aba with the one God of the universe. Thus the God
of Islam was engrafted onto a natural stock, which had its
root in the primitive Semitic mother goddess.

No doubt the few gods of Arabic heathenism which are
known to us through Mohammedan sources originated, at
least most of them, in the same way. Their names are epi-
thets, as Dhu-'l-Khalasa, Al-Fals, Al-Galsad, Al-Uqaisir,
etc. 2 They were each connected with some idol which

1 Op. cit., pp. 75, 76. Among the customs which Mohammedanism in-
herited from this old cult is that which requires the pilgrim to make a cir-
cuit seven times around the Qa'aba. He must lay aside his own clothes
and put on two pieces of cloth, one around the loins and the other over
the back, but in the more shameless days of heathenism it was done with-
out any clothing whatever. Cf. Zwemer's Arabia, the Cradle of Islam,
p. 38.

2Cf. Wellhausen's Heidentum, 2d ed., pp. 45-64.

TKANSFORMATIONS 135

was often a natural crag or stone. The details of their
history are unknown to us, but if we could ascertain them
we should no doubt find that these gods were developed
out of the primitive Semitic mother goddess by the same
laws of progress and differentiation, the action of which
we have already traced.

Passing now to Abyssinia, we find the worship of the
same deity, known here as Astar, the form which the name
of the primitive mother goddess assumed in Ethiopic.
The worship of Astar is vouched for in an inscription from
Aksum, fragmentary copies of which were brought to
Europe as long ago as 1833, and of which other fragments
have been secured at various times since. Bent in 1892
secured an almost perfect copy. The inscription was
written a little later than 400 A.D.^ There is also much
in Abyssinia besides the name to connect this deity with
the Athtar of south Arabia. At the site of the city of
Yeha, Bent found some fragmentary inscriptions which
made allusion to a place called Ava or Awa, apparently
situated on the present site of Yeha. Bent concludes that
Awa was the original name of Yeha. One of these frag-
mentary inscriptions reads "his house (temple), Awa."^
This inscription is written in the Sabeean script, evidently
by an immigrant from south Arabia. D. H. Miiller has
also pointed out that in two inscriptions from 'Amran (and
there are really more than two) Ilmaqqahu is called " Lord
of Awa™."^ Awa™ was probably a temple, since it was
situated in the city, Alw.* This confirms Bent's con-
jecture^ that Yeha was formerly called Awa, and was

1 Cf. D. H. Mailer's Epigraphische Denkmalern aus Abessinien, Wien,
1894, pp. 37, 38.

" Cf . Theodore Bent's Sacred City of the Ethiopians, pp. 145, 235, and
Mailer, op. cit; in n. 7, p. 61.

» Mailer, op. cit., p. 61, Bent, op. cit., p. 237. Cf. CIS., Pt. IV, Vol. I,
Nos. 74, 80, 99, 126, 147, and 155. Cf. Glaser's Die Abessinier in Arabia
und Afrika, pp. 103-105, and Hal^vy in Bevue semitique, Vol. IV, pp. 78-79.

* CIS., Pt. IV, Vol. I, No. 741

'Bent, op. cit., p. 145.

136 SEMITIC OKIGINS

founded by immigrants from south Arabia who brought
the cultus of that country with them. Ilmaqqahu, who is
described as " Lord of Awa™ " we have already found to
be a local development of Athtar, so that a chain of epi-
graphic evidence connects the worship in ancient Yeha
with the old Arabian mother goddess.

The deities of this cult in Arabia were often thought to
reside in a crag or stone,i one of which may still be seen
near Taif, in which Al-Lat was once thought to make her
home. 2 These stones corresponded to the massebas of the
northern Semites. The temple at Yeha still exhibits
traces of monoliths which answered a similar purpose, and
like those at Aksum, to be mentioned presently, help us
to identify the widely extended traces of this ancient
cult.3

The two countries on the opposite shores of the Red Sea
were at this time closely united, as the epigraphic evidence
shows,* and Glaser contends that Habashat, the old Semitic
name of which Abyssinia is a corruption, was not confined
to the African side of the sea, but designated a part of the
Arabian peninsula as well.^ It is, however, beyond dis-
pute, in consequence of the monumental evidence, that
migration to Africa took place after the masculine Athtar
had been developed, and that close political relations were
maintained between the two regions for a considerable
period of time.

Bent has shown that when hard pressed by their foes at
Yeha, the Semitic Abyssinians removed their capital to
Aksum and transferred their shrine thither. This shrine

^ Cf. Wellhausen, Heidentum, 2d ed., pp. 45-64, and Smith, Beligion
of the Semites, 2d ed,, pp. 201, 204, and 340.

" Cf. Doughty's Arabia Deserta, Vol. II, pp. 515 and 517.

'See Bent, op. cit., p. 139, and Wylde, Modern Abyssinia, pp. 150 ff.,
154 ff.

* Cf. D. H. Miiller, Epigraphische Denkmalern aus Abessinien, pp.
75-79, and Hehraica, Vol. X, p. 202 ff.

5 See his book. Die Ahessinier in Arabia und Afrika, passim, and
ZDMG., Vol. L, pp. 294, 295.

TRANSFORMATIONS 137

was marked by a large number of monoliths which served
as noshs or massebas, though like those at Yeha they were
somewhat more developed than those of the Israelites and
Cypriotes, since they had altars at their bases.^ They are
taller than the monoliths found by Bliss at Tell-es-Safi in
Palestine,^ but like them seem to have been massebas.
These objects, whether called nosbs or massebas, or by what-
ever name, were in general in the form of rude phalli, and
were no doubt chosen as the symbol of Semitic deity
because of their resemblance to the organ of the god of
life. Bent noted that the altars attached to the monoliths
at Aksum, as well as certain decorations which they bore,
were on the side of the rising sun.^ This, he inferred,
connected them with sun worship. The presence of the
name Astar at Aksum, together with the chain of evidence
which connects Athtar with Yeha, enables us to see in the
orientation of these monoliths an evidence of the worship
of Athtar Sharqan, with whom, as in Arabia, Shams was
probably associated. This inference is confirmed by the
traces of heathen ritual which have survived in the
Abyssinian church. All the church festivals are cele-
brated with music and dancing like heathen orgies.* On
entering the church the threshold and door posts are
kissed, showing that they are held to be sacred. ^ The
great festival of the year is the feast of the Cross, which
occurs in September, the month of the old Semitic date
harvest festival, and which we have already identified
with it.^ An important part of the celebration of this
festival is the building of fires on the high places and the
slaughter of oxen before sunrise, — traits not only heathen
in their origin, but which connect themselves in form with

1 Bent, op. cit., pp. 180, 182, and 185.

2 Cf. Pal. Expl. Fund's Quarterly Statement for October, 1899, pp.
317-320.

'Bent, op. cit., p. 190.

*Bent, ibid., pp. 53, 83, 84, and 165, also A Visit to Abyssinia, by W.
Winstanley, London, 1881, Vol. II, p. 127.

^Winstanley, op. cit., p. 127. 'Above, p. 112.

138 SEMITIC ORIGINS

the morning sacrifice of the camel to Al-Uzza by the Arabs
of Sinai, which the son of Nilus witnessed. ^ The autumn
festival of the old mother goddess is scarcely disguised by
its Christian name. The church at Aksum is also prob-
ably the old temple ; and to this day the old Semitic right
of asylum is enjoyed there by the wrong-doer^ as it was
in Israel at the altars of Yahwe.^

Abyssinia consists for the most part of a high tableland
on which crops are easily grown. The country is there-
fore an agricultural one, and subsistence is decidedly easier
than in Arabia.* The central and southern portions are
especially fertile.^ These conditions of life wrought, so
far as we can tell, no change in the Semitic family or the
Semitic conception of deity. The migration occurred so
late that the development traced in Arabia had already
taken place, and the cult was transplanted bodily to Africa.

The inscription of Ezana, which contains the name of
Astar, couples with it two other deities, Barras and Medr.
These gods are, I believe, not otherwise known, but it is
probable that one or both of them arose from Athtar, first
having been used as an epithet, and coming afterwards to
be regarded as a distinct god on account of its separate
name. In an inscription published by Halevy,* lUmaq-
qahu, whom we have already shown to be an Athtar, is
called " Lord of Medr." In this expression, Medr is the
name of a place.'' According to a passage in Iklil, quoted
by Sprenger,^ there stood opposite the mosque of Medr a
large castle with a marble slab on which was a picture of

1 Smith, Beligion of the Semites, 2d ed., pp. 166, 281, etc.

2 Bent, op. cit., p. 163.

8 Cf. Ex. 2112-w, and 1 Kgs. 1 and 2.

<Bent, op. cit., pp. 67, 79, 90, 91, 135, 136, 154, and 202, also Wylde,
op. cit., ch. xi.

5 Geology and Zoology of Abyssinia', by W. T. Blandford, The Mac-
millanCo., 1870, p. 196.

8 Journal asiatique, Ser. 6, Vol. XIX, p. 164, No. 172.

' Cf. Mordtmann and MiiUer's Sabdische Denkmdler, p. 59.

' Alte Geographie Arabiens, p. 221.

TRANSFORMATIONS 139

the sun and moon. It is conceivable that Athtar was
called " Lord of Medr " until the title was abbreviated
to Medr, who, when his devotees had removed to Africa,
was in time regarded as a god separate from Astar.

The case of Barras is even more obscure. D. H. Miiller
has conjectured^ that he was the god of thunder and
lightning because the Arabic barasa means to gleam or
flash (micare). If there be any value in this guess, the
name " Thunderer " may well have been an epithet of
Athtar. These are, however, only possibilities, and the
gods may have originated in ways quite different, — they
may, for example, have been native Abyssinian deities,
whose place was fixed before the Semitic immigration,
whose favor the Semites felt bound to propitiate, as the
Babylonians whom Sargon settled in Samaria felt bound
to propitiate Yahwe.^

Turning now to the countries north of Arabia, we come
first to the land of Moab. This country forms part of the
fertile strip at the eastern end of the Mediterranean which
we call Syria. This region is cut off from those sterile in-
fluences which render Arabia a desert by the proximity of
the sea and by its two ranges of mountains.^ A rainfall is
thus secured and the country redeemed from the encroach-
ments of the desert. The most easterly of the mountain
ranges of Syria forms the eastern bulwark of Moab toward
the desert, bringing its high plateaus within the region of
rain and fertility.* The elevated plains of Moab must
have been from time immemorial excellent pasture lands ;
and the few glimpses which the Old Testament and the
Moabite Stone give us of its industry, confirm us in the
belief that the Moabites were engaged mostly in pastoral
pursuits. Thus the tribute paid by Mesha to Omri and

1 Epigraphische Denkmaler aus Abessinien, p. 44.
» Of. 2 Kgs. 1724-34.

* See G. A. Smith's Historical Geography of the Holy Land, pp. 45-48.

* Cf. G. A. Smith, as above, and Hull's "Geology of Palestine" in
Hastings's Dictionary of the Bible.

140 SEMITIC ORIGINS

Ahab was paid in wool,^ while the sacrifices which Mesha
claims to have offered to his god consisted of sheep alone. ^

The conditions of life in a country like this must have
been far easier than in Arabia. How far back in antiquity
the Semites began to overrun Moab, no one knows. The
language of the Moabite Stone is practically identical
with the Hebrew of the Old Testament. A dialect
identical with both was spoken in Canaan in the fifteenth
century B.C., and has influenced the Babylonian of the
El-Amarna letters.^ In their traditions the Israelites also
recognize the Moabites as their kinsmen.* These facts,
together with the character of the Hebrew-Moabitish
language, which belongs to the north Semitic group, —
languages which have undergone a long development
independently of Arabic, — make it evident that the Moa-
bitish emigration was part of a movement which took
place many centuries before the date of the Moabite
Stone.

In a country like Moab, where the conditions of life are
not as severe as in Arabia, social evolution would proceed,
even if undisturbed by outside foreign influences, much
more rapidly than in the Arabian peninsula itself. Keas-
bey's researches into the formation of the clan^ have
shown that in the pastoral stage of civilization the patri-
archal clan is formed as the natural result of the environ-
ment. The conditions, therefore, in Moab must have
produced a patriarchal family centuries earlier than it
was produced in Arabia. The few references to Moab in
the Old Testament and the text of the Moabite Stone
confirm this view. The patriarchate was fully established
there before the ninth century B.C. How long before we

1 Cf. 2 Kgs. 3* and Moabite Stone {e.g. Smend and Socin's Inschrift
des Konigs Mesa von Moab, Freiburg, 1886), 11. 3-9.

2 Cf. Moabite Stone, 11. 30, 31. ip: and JSJ: are the terms used.

' Cf. Zimmern in Zeit. d. deutsch. Palastina-Vereins, Vol. XIII, H. 3,
and ZA., Vol. VI, pp. 245-263; also my "Peculiar Use of Iia.ni," in
PA08., 1892, p. cxcix.

♦ See Gen. 18 and 19. * See above, p. 30.

TRANSFORMATIONS 141

do not know. Their chief divinity, Chemosh, is a male
deity, — a fact which presupposes a patriarchal society for
a time sufficiently long to influence their religious con-
ceptions.^

Contrary to the opinion of many scholars, I believe
Chemosh to be genetically connected with the old Semitic
mother goddess. This opinion rests in part on the anal-
ogy of the south Arabian development already traced,
but largely on the interpretation of the name, Ashtar-
Chemosh of 1. 17 of Mesha's inscription. Baethgen,^
Driver,^ Moore,* and Peake,^ hold that this deity Ashtar
is not identical with Chemosh, but is an Ashtar, or Astarte,
who was associated in worship with him. Moore suggests
that it is parallel to Malik-Ashtart ^ and to the Ashtart
worshipped in the shrine of the god Hamman of the
Ma' sub inscription.^ This view appears to me untenable
on the following grounds : 1. The parallels urged are all
much later in date than the Mesha inscription. They
represent movements of thought influenced by Persian or
Greek ideas, or by both. The combination Ashtar-Che-
mosh may fittingly be compared with the combination
Yahwe-Elohim with which it is approximately contempo-
rary,* but not with Melek- Ashtart, which is considerably
later. Such comparison suggests the identification of god
with god on account of political union, but not the union
of a god with a goddess. 2. Ashtar in the inscription of
Mesha lacks the feminine termination, and is therefore

1 Cf. Moabite Stone, U. 1, 3, 5, 6, 9, 11, 12, and Nu. 2129, i Kgs. IV-^,
Jer. 48'- "• «, etc.

^ Beitrage zur semitischen Beligionsgeschichte, 1888, p. 14.

• Article " Ashtoreth," Hastings's Dictionary of the Bible, p. 171a.

* " Chemosh " in Encyc. Bib.

* " Chemosh " in Hastings's Dictionary of the Bible.

• CIS., Pt. I, Vol. I, No. 8.

' Cf. G. Hoffmann, TJber einige phoen. Ins. , p. 20.

' This statement is based, not on the form Yahwe-Elohim as it appears
in our present Biblical text, where it is made to appear to be a harmoniza-
tion with the late P document, but on the practical identification of the
two when the J and E documents were united about 650 b.c.

142 SEMITIC ORIGINS

a god and not a goddess. True, in primitive Semitic the
name designated a goddess without the help of a feminine
ending ; it is also true that in Babylonia and Assyria the
name continued to do so down to the latest times ; but
wherever the name has been found among the southera
Semites without the feminine termination, it designates
either an actual or a nascent god, and wherever it is found
among the western Semites designating a goddess it has
the feminine ending. It seems safe to conclude, there-
fore, that the name in Moab which was on the border of
Arabia and Canaan and intimately connected with the lat-
ter, would, when lacking a feminine termination, designate
a god. To break the force of this consideration one of two
things should be clearly proven, either that the feminine
ending was added to the name by the rest of the westera
Semites after the time of Mesha, or that Babylonian influ-
ence is responsible for its disuse here. Although all the
Biblical and Phoenician material containing the name is
later than the Moabite Stone, the name occurs twice ia
the El-Amarna tablets ^ as the name of a city, — no doubt
the Biblical place which bore the name "Ashtaroth,"
named from the goddess, — and here the feminine ending
appears. The west Semitic custom had attached the
feminine termination to the name by the fourteenth cen-
tury B.C., and that too on Moab's very border. The only
reason for suspecting Babylonian influence in Moab is the
possible connection of a proper name or two with Babylo-
nian names, such as Mount Nebo with the name of the Baby-
lonian god Nabu. But even if the name of a mountain
and a city survived from the time of Babylonian occupa-
tion (which is uncertain), that is insufficient ground for
supposing that the name, Ishtar, survived for six hundred
years without modification, when all the neighboring

1 Cf. KB., Vol. V, pp. 263 (No. 1421"), and 353 (No. 237"), also Gem.
146, and Josh. 13^1. For the identification of the localities mentioned ia
these letters, and the demonstration of their east Jordan situation, see
Sayoe's Patnarchal Palestine, pp. 133 ff. and 152 ff.

TRANSFORMATIONS 143

people, with whom the Moabites were intimately associ-
ated, used it with the feminine termination. 3. Mesha
equates Ashtar-Chemosh with Chemosh. He says (1. 14
ff.)» " And Chemosh said to me ' go and take Nebo against
Israel,' and I went by night and fought against it from
break of dawn till noon, and I took it and killed all of
them, seven thousand men and boys and the women and
girls and slave-girls, for I had made them harim to Ashtar-
Chemosh." Now it seems clear that the king would devote
his victims to the god who sent him forth to battle, — the
god who held, as the inscription shows throughout, the
same relation to the nation as a whole, which Yahwe bore
to Israel. Chemosh appears alone at the end of the in-
scription, 11. 32, 33. Ashtar-Chemosh cannot, therefore,
be even in part a different god from Chemosh. If under
such circumstances he had desired to associate a goddess
with Chemosh, he would hardly have put her before him.
It seems more natural to suppose that Ashtar-Chemosh
like Yahwe-Elohim is the union of two names into one
compound designation, either element of which might be
used for the god alone.

The course of religious development in Moab must,
therefore, have been not unlike that in south Arabia;
the mother goddess under the pressure of social transfor-
mation became a father god, and through the use of epi-
thets gradually came to be called by another name. This
development, as already observed, was earlier by centuries
than that in Arabia.

If this be true we should expect Chemosh to be a god of
fertility. There is even in the very scanty material extant,
some indication that this was the case. An old poem,
twice quoted in the Old Testament,^ makes the Moabites
his sons and daughters. He seems also to have been a
ba'al, or god of the land. I can see no good reason for
denying with Moore,^ a real identity between Chemosh

1 Nu. 21 27-3" and Jer. 48 «■ «.

* "Chemosh," in Encyc. Bib. Moore's denial probably is intended to

144 SEMITIC OEIGINS

and Baal Maoii,^ nor for his denial of the substantial
identity of Chemosh and Baal Peor (Nu. 25 3, Hos. 9^%
long ago perceived by Jerome. Baethgen is nearer the
truth when he regards both of them as forms of Baal.^
We do not need to insist that the people always thought
of the absolute identity of the god who was worshipped at
one shrine with the god who was worshipped at another,
any more than the untutored Catholic in modern Europe
always is conscious of the identity of the Virgins
adored at different shrines ; but the analogy of the Ath-
tars of south Arabia and of the Baals of Syria make it
practically certain that all had their root in the same god
of fertility, by whatsoever name each may have been called.

We learn from the prophet Hosea (91°), that the gross
practices characteristic of the worship of the Semitic
mother goddess in other localities, and which were so ab-
horrent to the prophets, were a part of the cult of this
Moabitish god. No doubt therefore as in south Arabia a
form of this goddess still existed in Moab side by side with
the male deity which had grown out of her, and that these
two, with slightly varying attributes at different shrines,
perpetuated in their worship the features which pertained
to this cult in other countries. This result remains even
if one were to disagree with the argument given above
for the sex of Ashtar in Moab ; for if Ashtar were a god-
dess, Chemosh, who is joined with her, would of necessity
be her male counterpart, or he could not be so joined. The
term ba'al would apply to him as aptly as we shall see that
it applied in Syria and Phoenicia. How the name Chemosh
came to be applied to this god we cannot tell ; it is still a
puzzle.

If now we turn to Phoenicia and Palestine, we come to
a land different in many ways from any of those hitherto
studied. Its contour is much more broken, and within

mean no more than that the gods were worshipped at different shrines
and therefore often thought of as practically distinct.

1 Moabite Stone, 1. .30. 2 Op. cit., p. 15.

TEANSFOKMATIONS 145

comparatively narrow limits there is greater variety of
scenery and climate than elsewhere in the Semitic world.
It is a maritime country, bordering as it does on the
Mediterranean coastland. Its two mountain ranges, —
the Lebanon, with its hills tapering off into southern
Judaea, and the range east of Jordan, — intercept the
moisture from the sea and give an abundant rainfall.
The deeply depressed valley of the Jordan and Dead Sea
affords a climate of tropical warmth, while the snow-capped
heights of Lebanon present the opposite extreme. ^ The
soil along the maritime plain and in the valleys of Esdrae-
lon and the Jordan is very fertile, while that of the hill-
sides is well adapted to the vine. Trees have always grown
in abundance on the hiUs, especially in the Lebanon region.
The Assyrian kings boast often that they took from here
beams with which to adorn their palaces.^ The low-lying
lands have always been well adapted to grain, while in
ancient times the palm tree grew in the Jordan valley and
along the sea coast. The olive and the vine were the chief
fruit bearers, the former being native in this region and
the latter probably so.^ The land has, in historical times,
been a land of orchards, the apricot, fig, pomegranate,
orange, citron, mulberry, pistachio, and almond being its
chief fruits, while the sycamore and carob tree yielded a
living for the very poor.* Wheat, barley, and many vege-
tables, such as onions, could be produced in abundance. In
such a land agriculture was born in the remote prehistoric
past, and its birth was inevitable. As George Adam Smith
remarks : ^ "To pass from the desert into Syria is to leave
the habits of the nomadic life for those of the agricultural.
The process may be gradual, and generally has been so,
but the end is inevitable."

•■ G. A. Smith's Historical Oeography of the Holy Land, ch. ii.
" Cf. e.g. KB., Vol. I, pp. 41, 108, and Vol. II, p. 113.
» 6. A. Smith, op. cit., p. 82.

* Cf. Amos 7" (Amos gathered sycamore figs), and Luke 15" (th«
prodigal son ate carob pods),
» Op. cit., p. 85.

146 SEMITIC ORIGINS

The beginnings of agricultural life in this region are
shrouded in obscurity. In the El-Amarna letters the
prince of Kumidi, near Gebal, sent, we are told, a tribute
of olive oil to the king of Egypt.^ The Egyptian monu-
ments of the middle empire tell the same tale. Wine,
figs, grain, and olive oil are mentioned as products of Syria
and Phoenicia,^ as well as several minerals, such as lead
and copper. Agriculture therefore must have antedated
in these lands the fifteenth century before Christ. Agri-
culture may have been preceded by the pastoral stage of
society, and the beginnings of the patriarchate and conse-
quently the general supremacy of male deities may date
from this stage of development.*

Of the coming of the Semites into this region we know
little. Lugalzaggisi, king of Erech 4000 B.C. or earlier,
whom Hilprecht believes to have been a Semite,* claims to
have subdued the country as far west as the Mediterra-
nean Sea.^ Sargon of Agade, about 3800 B.C., conquered
the Westland, or, as scholars generally regard it, the land
of the Amurru or Amorites on the north of Canaan.® A
contract tablet has been discovered which makes reference
to it in its date. From this time onward many Babylonian
kings claim to have conquered the Westland. That the
claim was real the El-Amarna tablets have proven by
showing that Babylonian culture had so penetrated the
country that the language and script of Babylonia had
become the regular vehicle of official communication. In
the fifteenth or sixteenth century B.C. the Eg3rptians over-

1 Cf. KB.,Yo\. V, p. 261. For location of Kumidi, cf. pp. 141, 187-189,
and 201.

2 Cf. W. Max Miiller's Asien und Europa, pp. 155 and 183.
' Hilderbrand's Becht und Sitte, pp. 31-34.

4 Cf. OBI., Pt. II, p. 54 ff.

6 Hilprecht, op. cit, p. 53.

6 Cf. Thureau Dangin in Comptes rendus de Vacademie dHnscriptions,
1896, p. 358 ff. (identical with his TaUettes chaldeennes inedites, No. 17),
also Driver in Hogarth's Authority and Archaeology, p. 40. For the
identiiication of Marlu (Westland), with Amurru, see Schrader in Sits-
ungsberichte of the Berlin Academy, 1894, p. 1301.

TRANSFORMATIONS 147

ran the region, and at a date which we cannot now deter-
mine the Phoenicians had made their way into it. The
Hittites also gained a strong foothold in the North earlier
than the Egyptian conquest. Great as the mixture of
races became, the Amorites appear to have maintained
their identity down to the Hebrew period. Their name
is still used by Amos and by the Elohist to designate the
old inhabitants of Palestine.^ This people who persisted
so long in a region where many races strove for the
supremacy had then assumed by 3800 B.C. an importance
so great that their land was coveted by the far-off Baby-
lonian. If they had not then developed agriculture, it is
difficult to understand why the country should have gained
such prominence, unless there were fisheries to make their
land conspicuous. There is some probability that the
agriculture which we are able to establish by monumental
evidence in the El-Amarna period is at least some twenty-
five hundred years older than that. The Amorites, like
other primitive peoples, must have had many local numina,
the most important of which would on the principles
already established be masculine deities.

What would happen when a band of Semites entered
this land, we may learn from 2 Kgs. 17^*- We are
informed there that the colony of Babylonians whom
Sargon settled in Samaria worshipped their own national
gods until such disasters overtook them that they felt
compelled to learn the worship of the god of the land.
The worship of Yahwe which was thus begun by them
did not cause them to forsake their old deities, but for a
time both were worshipped together, and at last a new
composite worship resulted. If one should object that
this occurred very late in Semitic history, the objection
would only strengthen our argument, for what could
happen at so late a date must a fortiori have happened in

1 Cf. Wellhausen, Jahrb. f. deut. TheoL, Vol. XXI, p. 602 ; Ed.
Meyer in ZATW-, Vol. I, pp. 121-127; and Stade, Geschichte, Vol. I,
p. 110.

148 SEMITIC ORIGINS

more primitive times. ^ Not altogether unlike this in
principle was the custom of the Aztecs in Mexico to sacri-
fice to the gods of conquered countries to propitiate them.^

At the period covered by the Old Testament and the
Phoenician inscriptions the chief god of each locality was
known as a ha'al, — a term which denotes the proprietor
or inhabitant of some favored place or district. Robert-
son Smith thought^ that among the Semites it designated
the divine proprietor of naturally irrigated land, and there
is much to be said for this view. Every city had its
Baal,* and there would seem to have been as many of
them as there were towns, cities, sanctuaries, or objects
which appeared to the worshippers to have a religious
significance. Thus there was the Baal of Tyre, the Baal
of Sidon, Baal-Hamman, Baal-Barith, Baal-Shamem, Baal-
Zebub, etc. In parts of Palestine this god was identified
with the sun and called Shemesh. The town of Beth-
Shemesh (1 Sam. 6) was named from his worship.

This worship of Baal was in many places connected
with the old mother goddess Ashtart; e.g. at Sidon an
Ashtart of the name of Baal is coupled with Baal as his
consort,^ and in the Old Testament Baal and Ashtoreth
are frequently classed together as though they belonged
to the same cult.^ It may be added that in North Africa,
which was colonized from Phoenicia, the mother goddess
Tanith, whom I have elsewhere shown to be an Ashtart,^
is constantly mentioned with Baal, and is sometimes called
" The Face of Baal. " ^ What connection has this worship
with the primitive conditions which we have discovered
for the Semitic stock ?

Moore suggests that these Baals were originally distinct

^ Cf . Budde, JReligion of Israel to the Exile, 1899, p. 54 fi.

2 Cf. Egville's Native Beligions of Mexico and Peru (Hibbert Lec-
tures, 1884), 2ded., p. 31.

3 Sel. of Sem., 2d ed., p. 97 ff. « Cf. CIS., Pt. I, Vol. I, No. 3".
■t Cf. Jer. 228, 1113. » gee, e.g., Jud. 10».

' Hebraica, Vol. X, pp. 48-53.

8 CtCIS., Ft. 1, Vol. I, Nos. 195, 263, and 380.

TRANSFORMATIONS 149

local numina.^ In one sense that is no doubt true, just as
the Ashtars of the primitive Semites were distinct numina,
though each numen in its oasis was formed under condi-
tions so similar to those which prevailed in other oases
that all of them possessed a common form.

When the Phoenicians or their ancestors first entered
Syria, it is clear that they brought the worship of their
Semitic goddess with them ; the survival of her worship
as an independent deity at Sidon and GebaP is sufficient
to prove this. We have shown that even in Arabia, as
society changed to the patriarchal type, this goddess was
transformed into a god, and it is clear that in an agricul-
tural country like Syria, already inhabited by a settled
and comparatively civilized people, which had for many
centuries been swept by wars of conquest,^ the process
would be greatly hastened. Each locality would have its
local god worshipped by the Amurru, or Amorites, or who-
ever the previous inhabitants were ; the incoming Semites,
like the Babylonians settled in Samaria* (2 Kgs. 17^^'),
would feel compelled to worship it ; they would at first
worship their own goddess also until in time her cult
would be blended in greater or less degree with that
of the god, as that of Yahwe came in time to be blended
in Israel with that of Baal.^ That this actually occurred
is shown by one of the El-Amarna tablets, which refers
to the goddess as " Ba'alat." * Thus syncretism helped the
progress of natural development, and made a male deity
supreme.

At Tyre the local Baal was called Melqart,'^ or "King

1 Article "Baal" in Encyc. Bib.
« See below, Chapter VI.

» For a discussion of the effects of war upon the sex of primitive agri-
cultural earth goddesses, see below, Chapter V.

* See above, p. 147.

' See Budde's Religion of Israel to the Sxile, Leot. II, and below,
Chapter VII.

• See KB., Vol. V, p. 139.

' Cf. CIS., Pt. I, Vol. 1, No. 122.

150 SEMITIC ORIGINS

of the city." It was doubtless the cult of this god which
the Tyrian princess Jezebel introduced into Israel.^ We
learn from Philo of Byblos that Astarte, Zeus Demarous,
and Adodos reigned over the countries, and that Astarte
took up her abode in Tyre.^ Zeus Demarous is probably
Melqart, while Adodos, or Hadad, does not belong to Tyre,
but is the Aramaean equivalent of Melqart. Thus at Tyre
a god and a goddess had developed in the cult, although
the persistency with which the old mother goddess clung
to her independence is shown by the fact that Philo still
names her first. From the Old Testament we learn, what
we should naturally expect, that in most Canaanitish towns
the Baal was the chief deity,^ and the same appears to
have been the case in most of the PhcEnician colonies, since
Baal is often addressed alone in their votive inscriptions.*
Perhaps the same was true of North Africa, though there
are some peculiar phenomena in the votive inscriptions
from that land. In the numerous cippi published in the
Corpus Inscriptionum Semitiearum, Tanith, though often
called " face of Baal," is usually mentioned before him.
The goddess is clearly subordinate to the god, but the
older Semitic feeling still leads the worshipper to place her
name first. The epithet, "face of Baal," probaby sur-
vived from a time of transition when both masculine and
feminine qualities were ascribed to the goddess, so that
she was represented with a female form and a bearded
face.^

At Sidon still a different development occurred. Sidon
had its Baal, to which was attached, as has been remarked,

1 See 1 Kgs. 18.

2 See Eusebius, Freparatio Evangelica, ed. Dindorf, I, 10, 31 ; Orelli's
Sanchoniathontis Fragmenta, p. 30 ; of. Hebraica, Vol. X, p. 31.

8 Cf. Jer. 1113.

* Cf. CIS., Pt. I, Vol. I, Nos. 12.3, 138, and 147, and the inscriptions
published by Philippe Berger, Actes du X7« cong. d. orent., Sec. IV, p.
273 ff.

6 See below, Chapter VI, and an article, "An Androgynous Babylonian
Divinity " in JOAS., Vol. XXP f., 186 fl.

TRANSFORMATIONS 151

an Ash tart "of the name of Baal"; but side by side with
this pair the mother goddess alone held her supremacy also,
for separate from the "Ashtart of the name of Baal" was
another Ashtart, whose priest king Tabnith was, and whose
priestess, his wife.^ This fact confirms the hint given in
the Old Testament phrase : "Ashtoreth, the abomination
of the Sidonians."^ In the midst of changes wrought by
syncretism and social transformation the worship of the
primitive goddess had survived in comparative purity at
Sidon, notwithstanding that in one phase she had been
subjected to Baal. We shall see later that this was not
an isolated phenomenon.

There are some curious combinations of divinities in the
Phoenician inscriptions, such as Melek-Ashtart, Eshmun-
Ashtart, and Eshmun-Melqart, but as I have pointed out
elsewhere,^ these are not primitive. They resulted from
influences which came into force in the West after contact
with Persians and Greeks.

We learn from Old Testament denunciations of Baalized*
Yahwe worship that Baal was worshipped on hill-tops, under
green trees, in spots marked by 'asheras, massebas, and
hammanim. Images were not always present, but when
there was a shrine the god was often represented by the
image of a bull. At his altars offerings of firstfruits and
firstlings were made ; and beside them fornication was not
only licensed, but consecrated. The god had priests who
leaped upon the altar and gashed themselves with knives,
and also a retinue of prophets. Of the connection of all
this with the worship of Yahwe we shall speak in a future
chapter. Similar organizations of the Baal cult existed
elsewhere. A fragmentary inscription attests a similar

1 Cf. CIS., Pt. I, Vol. I, No. 3 ; Bevue archeologique, July, 1887, p. 2 ;
and Hebraica, Vol. X, p. 29.

2 2 Kgs. 2313.

» In an article entitled " West Semitic Deities with Compound Namea "
in JBL., Vol. XX, pp. 22-27.
'Hos.2512.

152 SEMITIC ORIGINS

organization in Cyprus.^ Among the Edomites the god
seems to have been called Edom,^ and from his high place'
it seems that his worship was kindred to that of the Baalim.
Both he and the god Gad, for whom the tribe of Gad was
named,* probably at the bottom were Baalim, which orig-
inated like the others from the mother goddess.

The conditions in the north of Syria were not strikingly
dissimilar to those which prevailed in Phoenicia and Ca-
naan. The chief god of this region was called Hadad,
though other gods were not unknown.^ From the general
principles thus far established we should expect the origin
of Hadad to be not unlike that of a Baal. For reasons
which will appear as we proceed it will be better to post-
pone the discussion of this great Aramaean god until we
have passed in review the gods of Babylonia and Assyria.
The problems of these countries are, however, so complex
that they merit a chapter to themselves.

1 CIS^Tt. I, Vol. I, No. 86.
« Cf. W. R. Smith, Sel. of Sem., 2d ed., p. 42.

» Cf. Robinson's article, " The High Place at Petra," in Biblical World,
Vol. XVI, p. 66.

* Cf. Oriental Studies of the Oriental Club of Philadelphia, p. 108.

* See the Berlin Museum's Mittheilungen aus dem orientalische Samm-
Ittngen, Heft XI, p. 83.

CHRONOLOGICAL TABLE

163

CHRONOLOGICAL TABLE

Babylonian Chkonoloot

B.C.

E. A. Hoffman tablet and Father

Scheirs archaic texts, cir. 6000-

5500.
Blau monuments, cir. 5500-5000.
En-shag-kush-an-na, king of Kengi,

before 4500.
Urkagina, king of Shirpurla, cir.

4500.
Ur-Nina, king of Shirpurla, cir. 4300.
Eannadu I, king of Shirpurla, cir.

4150.
Entemena, Patesi of Shirpurla, cir.

4125.
Eannadu II, Patesi of Shirpurla, cir.

4100.
Lugalzaggisi of Gishban and Erech,

cir. 4000.
Lugaltarsi, king of Kish, cir. 3900.
Manishtuirba, king of Kish, cir. 3850.
Alu-usharshid, king of Kish, cir.3830.
Sargon, king of Agade, cir. 3800.
Naran-Sin, king of Agade, cir. 3750.
Ur-Bau, Patesi of Shirpurla, cir. 3200.
Gudea, Patesi of Shirpurla, cir. 3000.
Dynasties of Ur, Erech, Isin, and

Larsa, 3000-2400.
Ur-Gur, king of Ur, cir. 2500.
Dungi, king of Ur, cir. 2450.
First dynasty of Babylon, 2399-2094.
Khammurabi, king of Babylon, 2287-

22.32.
Second dynasty of Babylon, 2094-

cir. 1730.
Third dynasty of Babylon, cir. 1730-

1150.
Agum-kak-rimi, king of Babylon,

cir. 1700.
Kurigalzu II, king of Babylon, cir.

1300.

ASSTRIAM ChBONOLOOT
B.C.

lahmi-Dagan, Fates! of Assyria, cir.

1840.
Shamshi-Kamman, Patesi of Assyria,

cir. 1820.

Shalmeneser I, king of Assyria, cir.
1330-1300.

154

SEMITIC ORIGINS

CHRONOLOGICAL TABl,^— Continued

Babylonian Chronology

B.C.

Nebuchadnezzar I, king of Babylon,
1140 (founder of Pashi dynasty).

Nabu-apal-iddin, king of Babylon,
cir. 880.

Nabopolasser, king ol Babylon, 625-
604.

Nebuchadnezzar II, king of Babylon,
604-562.

Nabonidos,, king of Babylon, 555-
538.

Cyrus conquered Babylon, 538.

Babylon under the Persians, 538-331.

Babylon under Greeks, from 331
onward.

Inscription of Antiochus Soter (la-
test dated cuneiform inscription)
from between 280-260.

ASSYEIAN ChkONOLOGT
B.C.

Tiglath-pileser I, king of Assyria,

cir. 1120-1100.
Assur-nasir-pal, king of Assyria, 885-

860.
Shalmeneser II, king of Assyria, 860-

824.
Tiglath-pileser III, king of Assyria,

745-727.
Sargon, king of Assyria, 722-705.
Sennacherib, king of Assyria, 705-

681.
Esarhaddon, king of Assyria, 681-

668.
Assurbanlpal, king of Assyria, 668-

626.
Assyria conquered by Babylon, 606.

CHAPTER V

TEANSPOEMATIONS IN BABYLONIA

It is no easy task to apply the principles which have
been traced in the preceding pages to the phenomena of
the religion of Babylonia. The civilization of the Meso-
potamian valley is so old that its beginnings can only be
conjectured ; our information is so fragmentary concern-
ing the various periods of which we know something that
no complete history of the country can yet be written,
while the problem of its racial and linguistic origins is so
complicated that it has become the subject of heated con-
troversy. Notwithstanding all these obstacles, the princi-
ples of economic and social development can be applied
with considerable certainty, and by their application much
light is shed upon some of the complicated problems con-
nected with the genesis of Babylonian civilization.

The most ancient civilizations of the old world were
developed in the great river basins of the Nile, the Tigris-
Euphrates, the Ganges, and the Yang-tske rivers, where the
soil was rendered fertile by new material brought down by
the water.i The civilization of Babylonia was probably
the oldest of these. In the judgment of most Assyriolo-
gists we have written inscriptions from Babylonia dating
from a time as remote as 4500 B.c.,^ and it is probable that

1 Cf. The International Geography, ed. by Hugh Robert Mill, London,
1899, p. 436.

^ The writer holds ■with most Assyriologists that the statement of Nabo-
nidos {KB., Vol. Ill, Pt. 2, p. 105), that 3200 years elapsed between him
and Naram-Sin, the son of Sargon, may safely be taken as a working
hypothesis. Lehmann's acute suggestion in his Zwei Hauptprohleme d.
altorientalische Chronologie, p. 175 ff. (Leipzig, 1898), that it is a scribal

155

156 SEMITIC ORIGINS

the oldest picture writing is at least a thousand years older
tlian that, and a previous history of considerable length is
required for the development of this system of writing.

The beginnings of agricultural life in these regions can
only be conjectured. In far-off geologic time the Persian
Gulf extended far up toward the Mediterranean Sea.*
The whole valley of Mesopotamia has been gradually
formed, and in recent geologic time this has been done
largely by the detritus brought down by the rivers. About
seventy feet a year^ is added to the land in this way, or a
mile in seventy years. Both the Tigris and Euphrates have
annual periods of overflow on account of the melting of
the snow in the mountains of Armenia near their sources.
The Tigris begins to rise about the first of March, and the
Euphrates the middle of March ; the water of the former
is at its height in May and recedes in June or July, while
that of the latter rises till June, and not till September
has receded to its ordinary proportions.^ The soil has thus
been formed of rich materials, and the retreating flood
leaves it each year pulverized and well manured. There
is a considerable rainfall in November and December, and

error for 2200 is based largely on our Ignorance. Peters (PSBA., Vol.
VIII, p. 142) suggested that it was a round number, made up of an esti-
mate of eighty generations of forty years each. George A. Smith (Modern
Criticism and the Preaching of the Old Testament, p. 91, n.) holds the
eame view, and by reducing the generation to thirty-three years, would
fix Naram-Sin's date at 3190 b.c. The excavations now going on in the
East may falsify Lehmann's view any day. It will be time enough to
reduce Nabo-nidos's statement when more of the mounds have been forced
to relinquish their secrets, and it has been demonstrated that the ?aps in
our present knowledge cannot be filled. Cf. Rogers, History of Babylonia
and Assyria, p. 318 ff. If the oldest historical inscription is 4500 b c, the
Blau Monuments must date from at least 5000 B.C., and probably as early
as 5500 B.C. The E. A. Hoffman tablet and Father Schiel's would then
seem to be as old as 6000 b.c,

1 See the map in Geology, Chemical, Physical, and Stratigraphical, by
Joseph Prestwich, Oxford, Clarendon Press, 1886-8.

^ International Geography, p. 447.

' Rawlinson's Ancient Monarchies, Vol. I, p. 12, and Jastrow's Seli-
gion of Babylonia and Assyria, p. 29.

TRANSFORMATIONS IN BABYLONIA 157

then only occasional showers till May. Wheat and barley,
which were indigenous to this region, were probably culti-
vated first on the outskirts of the inundation, where the
soil had been naturally prepared for it. The rainy season
comes on just in time to give the grain a start after the
river floods have passed, and in the spring the harvest
occurs before the water attains its height. Here the
natural conditions combine to make agriculture easy, and
here there was in consequence developed one of the oldest,
if not the very oldest, agricultural communities of which
there is any record. Payne holds ^ that agriculture is
usually developed before a tribe is settled in the most
favorable position for husbandry, and that when they have
outgrown the resources of the spot where they became
agricultural they migrate to a more favorable environment,
where an opportunity is afforded to attain a higher civili-
zation and enter upon a grander history. It is perhaps
the case that this is true of the originators of Babylonian
agriculture, but there have been so many changes in
Babylonia that we cannot now speak with any certainty
on this point. No spot more suitable for the beginnings
of an agricultural life than Babylonia can well be imagined.
Wheat, barley, and sesame were no doubt the grains first
cultivated. They are indigenous to the region, and play
an important role in the many Babylonian contracts and
revenue lists which have come down to us, both those from
the dynasties of Ur, about 2500 B.C. ,2 and the numerous
contracts which come from the eighth to the fifth centuries

1 Sistory of America, Vol. II, p. 61.

" For instance of these grains see CTBM., Pt. I, No. Bu. 94-10-15, 4,
1. 1 ; Pt. II, No. Bu. 91-5-9 2178 A, 1. 1 ; Pt. Ill, No. 18343, Col. I, 1. 1 ;
No. 16368, obv. 1. 1 ; Pt. IV, No. Bu. 88-5-12, 504, 1. 1 ; Pt. VI, Nos. Bu.
91-5-9, 476, 1. 1, and 91-5-9, 2421, 1. 1 ; Pt. VII, Nos. 13160, 1. 1 ; 13318,
1. 1 ; 18376, 1. 1 ; 18395, 1. 1 ; 18.397 passim ; 18403, 1. 1 ; 18409, 1. 1 ;
18410, 1. 1 ; 18414, 1. 1 ; 18415, 1. 1 ; 18419, 1. 1 ; Pt. IX, Nos. 21386
passim ; 17748, 1. 1 ; 20007, 1. 1 ; Pt. X, 14-308, rev. Col. VIII, 1. 1 ; 21381,
rev. 1. 1 ; 18964 passim, etc., the tablets published and translated in
Radau's Early Babylonian Hixtory, pp. 418-433, also, Reisner's Tempel-
urkunden aus Telloh, Berlin, 1901, pp. 16, 137.

158 SEMITIC ORIGINS

B.c.^ These were always the most abundant grains; they
figure largely in the payment of taxes (which were often
paid in kind) and are among the most frequent subjects of
contract between individuals. Still earlier, in the time of
Sargon and Naram-Sin (3800-3700 B.C.), Agade was
noted for its perfect grain, and the grain of Agade was in
demand at the market of Shirpurla.^

Along with these grains there are lists of cattle, sheep,
asses, horns, hides, etc., which were given in payment of
taxes to the temples of Shirpurla and Ur.^ Pasturage was
therefore combined, as we should expect, with agriculture
in the economic life of ancient Babylonia. The fertile
valleys which led out from the great valley of the rivers
were admirably adapted to pasturage. Individual property
in land must have existed here several thousand years ago.
Estates were bought, sold, and rented, as the contract tab-
lets show, as early as 2300 B.C.,* and we have a plot of an
estate of complicated character and peculiar shape, which
dates from the fourth millennium before our era.^ An un-
published archaic tablet in New York, which probably

1 As an examjile of the evidence from the latter contracts, see the
numerous citations made to the Nabu-na'id contracts in Tallquist's
Sprache der Contracte Nahu-na'uls, pp. 130 and 138.

^ CI Thureau Dangin's Tablettes chaldeennes inedites, Paris, 1897,
Nos. 13, 1. 1 ; 29, 1. 1; 41, 1. 1 ; 43 passim, and p. 9 ff.

3 Cf. Thureau Dangin's Tablettes chaldeennes inedites, Nos. 12 : 35, 55,
rev. 1 and 70; CTBM, Pt. I, No. Bu. 94-10-15, 5, Col. I, 11. 1 and 26,
rev. Col. II, 1. 1; 94-10-16, 26, rev. Col. II, 1. 6; Pt. V, Nos. 12913
. passim ; 18993 passim ; 19024 passim ; Pt. VII, Nos. 12938, rev. 1. 13 ;
12944 pas.sm ,- 1S383 passim ; 18434,1. 11; 11766 passim ; 12939, Col.
II, 1. 18 ; 12929 passim ; 18382 passim ; Pt. IX, No. 19055 passim ; Pt.
X, Nos. 19064, rev. passim; 19772 passim, etc., Radau, Early Baby-
lonian History, pp. 354-409, where several tablets of the same character
are published and translated, and Reisner's Tempelurkunden aus Telloh,
pp. 24, 25, 26, 28, 30, 31, 33, etc.

* See the references in Meissner's Altababylonische Privatrecht, pp.
9-11.

^ Cf. Oppert's article, "Un cadastre chald^en du quatrigme miWnlum
avant I'fire chr^tienne" in the Comptes rendiis de V Academie des Inscrip-
tions et Belle-Lettres, 4 me ser, Vol. XXIV (1896), pp. 331 ff. ; and Thu-
reau Dangin's in Bevue d'' Assyriologic, Vol. l\', p. 13 ff.

TKANSFOEMATIONS IN BABYLONIA 159

dates about 6000 B.C., presents a plot of ground to a temple
and so proves that individual ownership of land even then
existed.

As we have learned in the preceding pages, the structure
of society in such a community would be in a sense patri-
archal. Kinship would naturally be reckoned through the
father, and this would as naturally find reflection in the
religion by making masculine deities prominent, if not by
placing masculine deities at the head of the pantheon.

As we have noted above,i the date-palm was a sacred
tree in Babylonia, but whether native there or whether
its culture was imported from Arabia has been a moot
point among scholars. ^ At present the palm is so abun-
dant in lower Mesopotamia that it is said that a proper
coat of arms for the country would be a date-palm. ^
Dates were, during the period from which most of our
contract tablets come, — the period from the eighth to the
fifth centuries before our era, — a staple article of diet
and of commerce.* We have not as yet so much evidence
of their commercial use at an earlier period, but they are
mentioned several times in the revenue lists of Gamil-Sin,
Bur-Sin, and other kings of the second dynasty of Ur.*
An interesting tablet from Telloh, dating from the time
of Naram-Sin (about 3750 B.C.), informs us that twenty-
six and one-half shekels of " dates of Agade " were re-
ceived at the city of Shirpurla.® It follows that dates
were cultivated at Agade in the first half of the fourth
millennium B.C. which had a sufficient reputation to be

1 P. 90 ff.

2 See the opinions cited above, Chapter II, p. 75, n. 4.
' Cf . Zwemer's Arabia, the Cradle of Islam, p. 121 .

* For examples of the abundant references in the contracts of this
period, cf. Tallquist's Sprache der Contracte Nahit-naHds, p. 111.

6 Cf. CTBM., Pt. Ill, No. 18958, rev. 11. 18, 22; Pt. VII, No. 17765
passim; Pt. IX, Nos. 17748, Col. II, 11. 10, 13; 19054 possir/i. Palm
tree wood is also mentioned, Pt. VII, No. 18.390, 1. 1.

* Cf. Thureau Dangin's Tablettes chaldeennes inedites, No. 48, Col. II,
1. 4, and also p. 9.

160 SEMITIC ORIGINS

distinguished from the dates of other places. That dates
were highly regarded at Shirpurla is further proven by
the fact that Entemena Patesi of that city, who lived
about 4100 B.C., built a house for the storage of dates, —
a fact of sufficient importance to be mentioned among his
titles to fame.^ That the palm was known long before
that in Babylonia is made probable by the presence of
a sign in Babylonian writing, which is perhaps derived
from the palm tree, and which occurs as early as the
inscription of Lugalzaggisi, about 4000 B.C., in a very
primitive form.^ It has the values sagr=damaqu = "to
favor," and ^m»imar="palm tree." Delitzsch holds ^
that the sign is composed of three elements, one meaning
"favor," one "people," and one "open" or "bestow";
and that because of its great usefulness the palm was
designated " (the tree which) is full of favor to men."
Ball on the other hand considers * that the sign is derived
from the application of the cone-like instrument borne
and applied to the tree, by the winged figures in the
Babylonian and Assyrian sculptures, and that it is there-
fore a picture of the fertilization of the date-palm. In
this case the idea of "favor" (sag') would become con-
nected with the sign on account of the peculiarly useful
function which the palm performed in ancient Babylonian
life. The older texts give the sign in a form which favors
Delitzsch's explanation rather than Ball's. This harmo-
nizes also better with other considerations concerning the
culture of the palm in Babylonia which we will now

1 Of. GTBM., Pt. X, No. 86900, 1. 14. The line reads E-TUR-RA KA-
LUM-MA MU-NA-RU, "A house for the accumulation (literally 'abyss,'
Briinnow's List, No. 10220) of dates he built." Cf. also Rev. d'Asst/r.,
Vol. II, 148, 149 ; De Sarzeo's Decouvertes, pi. 5, bis No. 1, a ; and Radau,
op. cit., p. 113, for a similar expression.

2 Cf. OBI., No. 87, Col. Ill, 11. 30, 32, and Thureau Dangin's Becher-
ches sur Vorigine Vecriture cnneiforme, No. 137. For the later form
and meanings, cf. Briinnow's List, p. 305.

3 Entstehung der altesten Schriftsystems, oder der Ursprung der Keil-
schriftzeichen, Leipzig, 1897, pp. 144, 145.

* Of. FSB A., Vol. XVI, p. 193.

TRANSFORMATIONS IN BABYLONIA 161

adduce. We accept the fact of the presence of the palm
and the use of the date at the period of Sargon and Naram-
Sin, but the fact should be noted that the use of the date
and the culture of the palm does not seem to have been
general at that time. Of course the fact that the dates
of Agade are especially mentioned does not prove that
the date was not cultivated elsewhere ; it does, however,
prove that it was especially cultivated at Agade, whose
kings are among the earliest kings to write in Semitic.
The fact that a king of Shirpurla built a house for the
accumulation of dates, somewhat as another built one for
the storage of cedar ,^ — a wood of foreign origin which
had to be brought from afar, — gives some ground for the
supposition that the fertilization of the date-palm so as to
make it produce more abundantly was a comparatively
new introduction into Babylonia, and perhaps not gen-
erally adopted.^ It is probable, as Hehn and Fischer con-
tend, that in prehistoric time the date-palm extended from
the Canaries to the Penjab,^ and its presence in Babylonia
in about 6000 B.C., is attested by its occurrence as a picto-
graph in an unpublished text in the E. A. Hoffman collec-
tion in New York, translated below, p. 213, n. 5, but it by no
means follows that the artificial fertilization of it would be-
come everywhere known at an equally early date.* Indeed,

1 Cf. an inscription of Eannadu (Eannatum'), published by Thureau
Dangin in Comptes rendus de VAcademie des Inscriptions, 1899, p. 348,
PI. II, Col. ii ; also the translation in Radau's Early Babylonian History,
p. 72 ff.

2 As we shall see below, it does not follow that artificial fertilization
of the date-palm was introduced at Agade in the time of Sargon into
Babylonia for the first time, but only that a strong wave of Semitic in-
fluence and Semitic culture helped the date culture of Agade at that time
to reach a point of preeminence over that of other places.

8 See Hehn's Culturpflanzen und Hausthif.re, 6th ed., p. 273; and
Fischer in Petermann's Mittheilungen, Ergangungsband XIV, No. 64,
p. 11.

* The fertilization of the palm in Mesopotamia is still performed by
hand. The tree is climbed and the pollen sprinkled over the flowers.
See Zwemer's Arabia, the Cradle of Islam, p. 123.

162 SEMITIC ORIGINS

we should expect that, in an agricultural country like Baby-
lonia, where grain was indigenous and easily cultivated,
that the development of agriculture would remove the
spur of necessity which in Arabia compelled men to re-
sort to artificial fertilization of the date-palm to support
life. This view is confirmed by a contract of the fifth
century B.C., which shows that the process of getting a
date orchard started in Babylonia was so expensive that
a man was willing to forego the rent of the land for sixty
years for the sake of having it done.^ These general con-
siderations lead us to believe that the process of fertilizing
the date-palm was introduced by the Semites from Arabia,
and that Arabian or Semitic civilization was characterized
by the influences of the date-palm culture, as the earliest
civilization of Babylonia was characterized by the more
ordinary agricultural pursuits. This conclusion involves
the consideration of some knotty problems to which we
must soon proceed, and it will be found, when these are
considered, that several other considerations will confirm
the point of view here taken.

In such an agricultural country villages grow up in
protected centres where fortification is possible and where
it is accordingly possible to protect the growing crops
from the forays of more barbarous tribes. This was the
case in Mexico and Peru,^ in Egpyt, and was also no
doubt the origin of the Babylonian cities.^ These cities
were in the first instance the residence of fellow-tribes-
men and were built around the temple of their divinity
of fertility. All this in the development of Babylonia
lies in the prehistoric period. In that period, however,
Nippur, Eridu, Ur, Shirpurla, Kutha, Erech, Agade, and
other cities had sprung into existence. Before the dawn

' Cf. Hilprecht and Clay's Business Documents of MurashS. Sons of
Nippur, No. 48. Cf. ibid., p. 36 ff., and Assyrian and Babylonian Liter-
ature, N. Y., Appleton, 1901, p. 260 fe.

^ Cf . Payne's History of America, Vol. II, p. 47.

' See Winckler's Altorientalische Forsclmngen, Heft III, p. 232 ff.

TRANSFORMATIONS IN BABYLONIA 163

of OUT present historical knowledge, about 4500 B.C., the
struggle between these cities for supremacy had not only-
been begun, but had been waged with such varying for-
tunes that now one city had been supreme in power over
the others for a century or two, and now another. This
struggle, with its varied results, — Shirpurla being in
possession of empire for a time, then Erech, then Agade,
then Erech, Ur, Isin, Ur, and Larsa in succession, — con-
tinued until terminated by the final supremacy of Babylon,
about 2300 b.c.^ As will appear from arguments to be
adduced later, Nippur must have held the supremacy for
a long time during the prehistoric period. The political
combinations which resulted produced religious syncretism.
The city which was fortunate enough to win the leader-
ship for a few centuries would gain a high position for
its god in the minds of the inhabitants of the subjugated
cities, and the city which was sufficiently fortunate to
gain the supremacy first and to hold it for a long period
would win for its god the distinction of being the head
of the pantheon. That Nippur first held such empire the
position of its god Enlil (Bel) indisputably proves, and a
few fragments of archaic inscriptions attest.^ The gods
and goddesses of the other cities were grouped around
him as sons and daughters or in some other subordinate
position.3 Enlil, who held this position for two thousand
years, from the dawn of history to the rise of Babylon,
was finally displaced by Marduk, the god of the latter
city. How long in prehistoric time this process had
been going on we can only estimate. Our task is ren-

1 On this period of Babylonian history cf . Meyer, Geschichte des Alter-
thiims, Vol. I, 1884 ; Tiele, Bahylonische-assyrisch Geschichte, 1886-8 ;
Murdter-Delitzsch, Geschichte Babyloniens und Assyriens, 2d. ed., 1891 ;
Winokler, Geschichte Babyloniens und Assyriens; MoCurdy, History,
Prophecy, and the Monuments, 1894 ; Rogers, Outlines of the History of
Early Babylonia, 1895 ; Radau, Early Babylonian History, 1900 ; and
Rogers, History of Babylonia and Assyria, Bk. II.

2 Cf. OBI., Nos. 90-92, 94, 96, and 111 ; also PI. XVI.

8 Cf. Jastrow's Beligion of Babylonia and Assyria, chs. iii-vi.

164 SEMITIC ORIGINS

dered difficult by the fact that the beginnings which
we are seeking are not onl}^, as in other cases, shrouded
in prehistoric darkness, but that the traces of them
which can for the most part be detected in other parts
of the Semitic world were here very largely swept
away before the dawn of history by political and re-
ligious syncretism.

The difficulty of the problem is increased by the linguis-
tic and paleographic phenomena. As is well known, the
cuneiform inscriptions contain what most scholars regard
as two distinct languages, the Sumerian and the Semitic
Babylonian. It is generally held that the Sumerians
invented the cuneiform system of writing. Halevy first,
in 1874,1 and with much persistence in several publica-
tions since,^ has maintained that the so-called Sumerian
was only an allographic way of writing Semitic, and that
the Semites invented the cuneiform system of writing.
Guyard,^ McCurdy,* Price,^ Jeremias,® and Thureau Dan-
gin^ have come over to his theory, and though Delitzsch
had in 1889,8 i^ i896,9 and 1897,1° he had returned to
his former Sumerian point of view. The Sumerian theory
is based on the fact that there exist bilingual syllabaries

1 See Journal asiatique, 7th ser., Vol. Ill, p. 461 fi.

2 For a list of them cf. Weissbach's Sumerische Frage, Leipzig, 1898,
p. 25 ff. Halgvy's most complete grammatical statement of his point of
view is in Actes du sixieme Congres International des Orientalistes, Pts.
I and II, pp. 535-568. His latest statement is contained in a series of
articles in the Revue semitique for 1900.

' Cf. Bevue critique, nouv. ser., Vol. IX (1880), p. 425 ff ; and Sevue
de Vhistoire des religions, Vol. V, pp. 252-278.

* Cf. Presbyterian and Beformed Beview, Vol. II, 1891, p. 58 ff.; and
History, Prophecy, and the Monuments, Vol. I, pp. 87-95.

5 Ct. " Accadians," in Hastings's Dictionary of the Bible.

^ Cf. Chantepie de la Saussaye's Lehrbuch der Beligionsgeschichte,
Vol. I, p. 165 ff.

' Cf. Bevue d^assyriologie. Vol. IV, p. 73 fl. ; Tablettes chaldeennes
inedites, pp. 1-18.

' Cf. his Assyrian Grammar, pp. 61-71.

° Cf. his Assyrisches Handworterbuch, passim.

1° Cf. his Entstehung des ixUesten Schriftsystems, passim.

TRANSFORMATIONS IN BABYLONIA 165

and word lists,^ bilingual hymns and prayers,^ bilingual
inscriptions of kings,^ besides many unilingual inscriptions
in both languages.* The language of the portion of these
documents, called Sumerian, is held by most Assyriologists
as conclusive evidence of the existence of a non-Semitic
people, who gave birth to the language and invented the
script. Hal^vy contends that this was only a priestly
method of writing so that the uninitiated should not be
able to read it, that the syllabic values are all of Semitic
derivation,^ and that the Babylonian syllabary is perfectly
adapted to express the sounds of a Semitic language.
Delitzsch held, in 1889,^ that 106 signs were demonstrably
of Semitic derivation. To this number McCurdy has
added about forty more.^ The scholars of this school
also urge, that the fact that the Semitic inscriptions occur
side by side with the Sumerian back to 3800 B.C., together
with the fact that no Sumerians are mentioned in the his-
torical inscriptions, as the Elamites, Kossseans, etc., are,
is evidence that no such people existed.

The arguments of these scholars are persuasive, but not
quite convincing. We may grant the force of the fact,
that such texts as the prayer of Samassumukin ^ is in-
fluenced by Semitic idiom, and that a number of Semitic

1 See for example those published in II R., and in CTBM., Pts. XI
and XII.

2 See those published in IV R., and in Haupt's^^SJETT'., and by Reisner
in the Mittheilungen of the Berlin Museum, Heft X. Cf. also Zimmern's
Bnhylonische jBusspsalmen.

3 As, for example, that of Khammurabi. Cf. KB., Vol. IIP, p. 110 ff.

* The many royal annals of the Assyrian kings {KB., Vols. I and II)
may be cited as Semitic examples, while those of the kings of Shirpurla,
published in De Sarzec's Decouvertes en Chaldee, are examples of the Su-
merian.

' See Part 3 of his "Nouvelles considerations sur le syllabaire cunei-
forme," Journal asiatique, 7th ser., Vol. VII, p. 201 ff.

^ Cf. his Asayrinn Orammar, § 25.

' Cf. Presbyterian and Beformed Eeview, 1891, p. 58 ff.

' VB., 62, No. 2, and Lehmann's Samassumukin, Tafeln I and II.
Cf. Ft. II, p. 6 ff.

166 SEMITIC ORIGINS

idioms are found in Sumerian texts even back to the oldest
inscriptions,^ but there are a number of phenomena which
are not satisfactorily explained by the arguments of the
Halevy school. No satisfactory Semitic origin has as yet
been proposed for a considerable number of the oldest and
most common signs.^ The way in which Semitic words
have to be torn apart, in order to be expressed in the
cuneiform script, is hardly consistent with the supposition
that it was the invention of a Semitic people. The pecul-
iar verb prefixes and suffixes, the postpositions instead of
prepositions, and the various phenomena of the Sumerian
grammar, can by no process of argumentation be made to
appear the phenomena of a Semitic language, or the prob-
able invention of a Semitic people. There are not want-
ing, morever, in the bilingual texts instances in which the
Semitic idiom is so peculiarly modified that no explana-
tion of it seems adequate, except that it has resulted from
the influence of the idiom of the foreign language, of
which it is a translation.^ When we reflect, too, that
most of the oldest inscriptions are written in what Halevy
calls the allographic, or hieratic form, we are not only
confronted with the difficulty, to which Radau has called
attention* (viz., that the existence of this double form of
writing, as early as 3800 B.C., presupposes an incredibly
long anterior cultural development), but are compelled to

1 Cf. Hilprecht, OBI., Pt. II, p. 55 ; and Eadau, Early Babylonian
History, pp. 145-147.

2 The value an, for example, can hardly have originated, as Delitzsoh
{Grammar, p. 65) would have it, from a Semitic source. Nor can the
following be assigned to a Semitic origin : ud (utu), us, sal, sm, pi, du,
ha, ur, kur, gir, bu, and many others. In general, the signs which were
originally pictographs have values which cannot he explained on a Se-
mitic basis. Cf . my Studies in the Origin and Development of the Cunei-
form Syllabary, in preparation.

^ Such, for example, is the phrase, i-sap-pu-ru-su-nu = "they cry out "
(IV R., 1, Col. I., 1. 15), where sM-raJt is, contrary to Assyrian idiom,
the subject of i-sap-pu-ru. It is a literal translation of the Sumerian
O TJ-BAL-BAL-A-MES, which stands in the preceding line.

■* Early Babylonian History, p. 148.

TRANSFORMATIONS IN BABYLONIA 167

suppose, that such kings as Eannadu and Lugalzaggisi,
who wrote inscriptions to perpetuate their fame, chose to
have them written in a form which only a few could under-
stand. One can hardly believe, as hp would thus be com- / jv-»-
pelled to do, that the French bon mot, that language was
invented to conceal thought, was thus anticipated by these
kings at the very dawn of history.

Radau's view ^ is the one which my own studies had led
me to adopt, viz., that the Sumerians were the pre-
Semitic inhabitants of Babylonia, that they invented the
cuneiform system of writing, but that the Semites had
entered Babylonia and conquered them before the dawn
of history. The situation with which we are confronted
in early Babylonia is not altogether unlike that which
existed in Palestine in the period from which the El-
Amarna tablets come. In the latter country the Canaan-
ites had their own language, but had as yet no method of
writing it. The Babylonians had long dominated the
country, and their system of writing was well known.
To express themselves in written form, therefore, the
Canaanites had recourse to the Babylonian language and
script, though the Babylonians as a power in Palestine
had ceased to be for so long a time that no reference is
made to them in the El-Amarna, Palestinian letters. In
thus using Babylonian the Canaanites mingled their own
idioms with those of the foreign tongue.^ Similarly,
at the dawn of history the Semites had broken the
Sumerian power so long before that we find no mention
of the Sumerians in the inscriptions of Babylonia, though
to express themselves in writing the Semites were at first
compelled to resort to the Sumerian language and script.
In using these, however, they, like the Canaanites, mingled
their own idioms.^ In some important respects there is

1 Ibid., p. 149.

2 For examples see my article " A Peculiar Use of TlSni in the Tablets
from El-Amarna" in PAOS., 1892, p. cxcvi ff., especially p. ccxix.

" See the references given above, p. 166, n. 1.

168 SEMITIC ORIGINS

no parallel between Babylonia and the Palestine of the
period cited. For example, in Babylonia the Semites who
did the borrowing were the invaders, while in Palestine
the invaders were the people who furnished the script;
but the analogy holds for the important point to which
we have applied it, and helps us to understand the silence
of the inscriptions with reference to the Sumerians.^

The linguistic may be reenforced by other considera-
tions. While there are few elements of the Babylonian
religion which cannot be explained as Semitic, if one may
be permitted to draw analogies from agricultural Semites
outside of Arabia, yet there are some features which can-
not be so explained. For example, the early kings of
Babylonia were frequently deified. Even in their lifetime
their names were written with the determinative for deity
before them. Naram-Sin calls himself "god of Agade,"
and votive inscriptions are offered to other kings as gods ; *
while Gudea provided that certain sacrifices should be
offered, apparently to his statue, which was erected in the
temple of Ningirsu.^ Radau has tried to trace the growth
of this custom,* and finds it incipient in the inscription
of Lugalzaggisi, full grown in those of Sargon of Agade,

1 The arguments for the existence of the Sumerian language are forci-
bly stated from the older point of view in Haupt's Die sumerischen
Familiengesetze, 1879 ; "XJeber einen Dialekt der sumerischen Sprache,"
in Nachrichten d. K. Gesell. d. Wiss. zu Gottingen, 1880, pp. 513-542 ;
and ASKT., pp. 134-220 ; Sohrader's " Zur Frage naoh dem Ursprung der
habylonisohen Kultur" in ZDMG., Vol. XXIX (1876), pp. 1-52; and
Tiele's Geschichte, Vol. I, p. 68 ff. For more recent statements of the
argument see Lehmann's SamaUumukin, Ft. I, pp. 57 ff., 107 ff., and
Weissbach's Sumerische Frage, p. 150 ff. The view of the problem
taken in the text supposes that the Semites began to use the cuneiform
system of writing at a time so early that they exercised a large influence
on its later development. Thus the fact that the syllabary contains a
sign for Aleph and for other derivatives from sources demonstrably
Semitic is fully accounted for.

2 Cf. Radau, Early Babylonian History, pp. 164-166, 240, n. 1, 247,
250, 251, and 308 ff.

«Cf. KB., Vol. IIP, p. 27.
* Op. cit., p. 308 ff.

TRANSFORMATIONS IN BABYLONIA 169

and most flourishing in those of Naram-Sin, though it
persisted long afterward, as, for example, in the inscrip-
tions of the second dynasty of Ur. On the basis of these
facts he builds the theory that it was a point of view
characteristic of the Semites, and that Sargon, represent-
ing a Semitic migration from Arabia, had revived a
Semitic custom. ^ To argue thus is to erect a pyramid on
its apex. There is no Semitic analogy elsewhere for the
deification of kings, either during their lifetime or after-
ward. All we know of the culture of Arabia affords no
basis whatever for the view that such a custom could
originate there. The simple life of the desert and the
oasis threw men too closely together for even a sheik to
become a god to his fellow-clansmen. If there is a reli-
gious idea which we can pronounce absolutely un-Semitic
it is this. Thureau Dangin seems to recognize this when
he suggests that Egyptian influence led to the deification
of themselves on the part of Babylonian kings. ^ As the
empire of Naram-Sin extended to Palestine, he thinks con-
tact with Egypt may have occurred in a way to account
for the introduction of this practice. Such influence is
not intrinsically probable, and if it were, one wonders
why in later ages, under Kallima-Sin, Burnaburiash, Esar-
haddon, Assurbanipal, and Nebuchadnezzar, when contact
with Egypt was close and prolonged, no such consequence
of Egyptian influence resulted. In fact, such influence is
inadequate to explain the phenomenon. It must have
been an influence local, intimate, and prolonged, an influ-
ence from a non-Semitic source, but which after a few
centuries failed to be felt. It is just such an influence as
the Sumerian must have been. Radau seems to think
that because it first manifests itself fully at Agade that it
cannot be Sumerian,^ but do we know enough of the
habitat of these prehistoric people to be sure of this? I

J Op. cit., p. 310.

" Cf. Eeceuil de traveaux. Vol. XIX, p. 187.

» Op. cit., p. 309.

170 SEMITIC ORIGINS

think not. Moreover, evidence will be adduced below to
show that the Sumerian power was dominant in the North
rather than in the South. The presence of a non-Semitic
race in ancient Babylonia is further indicated by the faces
pictured on the votive tablet of Ur-Enlil at Nippur. ^
Professor Cope recognized in these peculiar faces the
Semitic nose, but a jaw which he regarded as Aryan. ^
He thus bears witness to the existence of a hybrid popu-
lation in this region at the dawn of history. The Blau
Monuments, which are still older, bear witness, as Ward
has pointed out, to the presence of two races in Babylonia.^
This is all we desire. We must confess that the Mon-
golian affinities of the Sumerians have never been clearly
proven. It is enough for our present purpose to show
that there was a mixture of races in Babylonia at this
period, and to agree to call the non-Semitic portion Su-
merian until such time as we can obtain a better name.

Another feature of the civilization of Babylonia points
to such a mixture of races. The decimal system of num-
bers was the native Hamito-Semitic system. Either it or
the quintal system, based on the fingers of one hand,* is
universally present among the Hamites and Semites, and
in Babylonia finally prevailed over the sexigesimal system
which was used in the earlier inscriptions.^ In Babylonia
the day and night were divided into six equal parts® — a

1 OBI., PI. XVI.

2 Cf. OBI., Pt. II, p. 48, n. 1.

' See American Journal of Archceology, 1st ser. , Pis. IV and V, and
p. 40.

* McGee's argument (American Anthropologist, New Series, Vol. I
[1899], pp. 646-674), although it adduces considerable proof in favor of
the influence of mythical or superstitious ideas in giving prominence to
certain numbers, really offers no explanation for the adoption of a quintal
or decimal system of numbers. The time-honored suggestion vyhich is
repeated in the text therefore seems valid.

5 Cf. the article "Number" in Encyc. Bib., by the writer.

^ See the interesting astronomical report published in III R., 51, which
was made at the time of a vernal equinox. It reads: "3d day of the
month Nisan ; the day and the night were equal. 6 Kasbu was the day ;

TRANSFORMATIONS IN BABYLONIA 171

measurement which, as Ihering has pointed out, cannot
have originated with a people who knew the decimal sys-
tem.^ We cannot go astray, therefore, in attributing the
invention of this system of numbers to the Sumerians who
invented also the cuneiform system of writing.

As we find in Babylonia convincing proof of the exist-
ence before the coming of the Semites of a non-Semitic
people who possessed a high degree of civilization, and
from whom the Semites borrowed the elements of their
system of writing, we have next to inquire whether there
is any test which we can apply to Babylonian religious
institutions which will enable us in any degree to distin-
guish its Semitic from its Sumerian elements. "We have
seen that the characteristic elements of primitive Semitic
religion are those produced in the desert and by the oasis
culture of the date-palm. There the feminine element of
society held a most important place, and in the religion it
was deified. In Babylonia, on the other hand, the eco-
nomic conditions were such that agriculture flourished
from time immemorial. A fertile and almost inexhaus-
tible soil yielded its riches to the husbandman. The date-
palm grew wild, and no doubt the fruit which it happened
to yield was gladly used ; but in all probability it was so
easy to raise grains that the pressure which compelled the
Semite in Arabia to cultivate the date-palm was lacking
in Babylonia. Fischer is probably right in claiming that
its culture, in the proper sense of that term, was intro-
duced from Arabia.^ In such a fertile environment every
man can obtain enough to support a wife, especially as in
early communities the woman performs much of the labor.
Monogamy is in such communities the rule for the com-
mon people, while polygamy is practised by the rich, of
whom such a community soon produces a considerable

6 Kasbn was the night. May Nahu and Marduk to the king my lord be
gracious ! "

1 See Ihering's Evolution of the Aryan, p. 121 ff.

" Cf. Petermann's Mittheilungen, Erganzungsband, XIV, No. 64, p. 11.

172 SEMITIC ORIGINS

number. 1 If now we can determine what kind of a reli-
gion the civilization of such a country would produce,
some light will be thrown upon our problem.

In search of an analogy which will supply our needed
clew, we may most profitably turn to the civilizations of
ancient Mexico and Peru, — two countries quite isolated
from the civilizations of the Eastern hemisphere and from
each other, but both of which produced civilizations and
religions of a high degree of organization. ^

Mexico consists of an elevated tableland, the surface of
which is covered with lava discharged from its volcanic
mountains, and of detritus which the storms of countless
centuries have washed down from its lofty ranges and
peaks. For the most part it has few rivers, but the region
around the city of Mexico is well watered and contains
many lakes. ^ The water supply and the subtropical cli-
mate in this part make agriculture easy. Here the an-
cient Mexican tribes developed their civilization. In its
completed form it was of a mixed character with Aztec
elements in the ascendency ; but in its earlier phases
simple agricultural communities, especially devoted to the
worship of their earth goddesses, or, as they often re-
garded them, maize goddesses, formed the earliest nuclei.*
These earth goddesses were, as Payne ^ has pointed out,
especially connected in their development with agricul-
ture. They were the original deities of the Totoncas, the
Otomi, and the Toltecs; and also of the Aztecs in tbeir

1 Cf. Payne's History of America, Vol. II, pp. 13, 15.

2 We cannot seek the analogy which we desire in Egypt because, as
we have seen, the religion of Egypt has back of it some elements of the
oasis civilization. Nor can we turn to India, China, or Japan with any
success because there the primitive agricultural religious product has lung
been displaced by more philosophical systems, or tortured by them into
unrecognizable forms.

' T%e, International Geography, p. 776.

*■ Cf. Payne's History of America, Vol. I, pp. 462, 516, 520, and VoL
U, p. 480.

^Cf. Payne, op. cit., Vol. I, p. 518.

TRANSFORMATIONS IN BABYLONIA 173

primitive northern home before they migrated to the South
and became conquerors. -"^

The civilization resulting from this Aztec conquest was
in many respects quite advanced. Property was organ-
ized for the nobles on a hereditary basis, and for the com-
mon people on a communal basis. Taxes were raised in
kind, as was the case in good degree in Babylonia.

Slaves, as in the latter country, did the laborious work.
Unlike the Babylonians, the Mexicans had no beasts of
burden and did not know the use of iron ; only gold, silver,
copper, and stone. ^ There were good markets and abun-
dance of wealth. In the cities associations of merchants
exercised great political influence. As in Babylonia,
polygamy was practised by the rich and by kings. The
Aztec emperor is said to have had a thousand wives. ^

In certain respects the resulting Aztec religion resem-
bled the Babylonian. The assimilation of conquered
tribes with the conquerors created religious syncretism,
and led to the formation of a pantheon. The form of
their temple, though much broader, bears considerable
resemblance to a Babylonian ziggurat.* The Aztecs, too,
like the Semites, thought that sacrifice united the wor-
shipper to his god.^ At the time when Europeans came
into contact with the Aztecs, tribe had conquered tribe
till much religious syncretism had resulted, a pantheon
had been organized, and as in Babylonia the functions of
the various gods had been much specialized. The heads

iCf. Payne, ibid., p. 520. For the Otomi, of. Payne, op. cit., Vol. II,
p. 454 flf.

^ Cf. The native Beligions of Mexico and Peru, Hibbert Lectures,
18H4, by Albert RSville, 2d ed., London, 1895, pp. 32, 33.

8 Mexico, Aztec, Spanish, and Bepublican, by Brantz Mayer, Hart-
ford, 1853, Vol. I, p. 36.

* Cf. The History of Mexico and its Wars, by John Frost, New Or-
leans, 1882, Vol. II, p. 40; Brantz Mayer, op. cit.. Vol. I, p. 37, and
Perrot and Chipiez, Histoire de Vart en Chaldee et Assyrie, Vol. II, p.
408 ff.

s Cf. Rgville, op. cit., p. 89, and W. R. Smith's Beligion of the Semites,
Lect., VI-IX.

174 SEMITIC ORIGINS

of the pantheon were identified with the sun and moon,
and were called grandfather and grandmother, i These
were in theory the chief deities, but in practice those were
worshipped more which stood nearer the interests of every-
day life.^ There was a wind god, usually pictured under
the form of a serpent,^ a form which had survived from a
previous condition of totemism. Tlaloc was the god of
rain — the god of fecundity to whom many children were
sacrificed.* Tlazolteotl was the goddess of love and
sensuality. Originally the wife of Tlaloc, the rain god,
the sun had stolen her away.^ Centeotl was the goddess
of the maize ; she had a son who bore the same name as
herself. She was often represented with this son as a
child in her arms, and reminded the Spaniards of the Ma-
donna and the child Jesus. ^ In addition to these princi-
pal deities they also had little household gods somewhat
like the Hebrew teraphim.^

It will be noted from this brief description that the re-
sulting Mexican civilization possessed both gods and god-
desses as did the Babylonian, and that, as in the latter
country, these were arranged in pairs. The goddesses who
had survived from the more primitive period were not,
however, supreme, but were subordinate to the gods. There
is a story to the effect that Tlazolteotl, the goddess of sen-
suality, prevailed over the pious hermit Yappan when he
had resisted all other temptations,^ as Ukhat prevailed
over Eabani;^ but Tlazolteotl was herself subject to her
divine husband, and was not supreme as was Ishtar.

The conception of the relation between gods and god-
desses is reflected in the procedure connected with their
great annual human sacrifice. The man chosen for this
sacrifice was treated for a year previous to its occurrence

iRSville, op. cit., p. 35. ^Rgville, op. at., p. 75.

2 Rgville, op. cit, p. 68. « Rfeville, ihUl, p. 73.

" Cf. Rfiville, op. cit, pp. 38, 68. ' Rgville, ibid., p. 77.

* RSville, op. cit., p. 71. 8 RgyiUe, md., p. 76.

" See above, p. 83.

TKANSFORMATIONS IN BABYLONIA 175

with divine honors. For at least a month before the sacri-
fice took place four beautiful girls were given him to share
his bed, and he passed his time in dalliance with these
until the day of the sacrifice came around. ^ It thus ap-
pears that the chief deity to whom this sacrifice was offered
was conceived as a polygamous god, the possessor of a
harem of goddesses.

The story of the seduction of Yappan bears, therefore,
only a superficial resemblance to the story of Eabani.
There is no trace among the Aztecs of a general worship
of deified polyandry or unwedded love as among the Sem-
ites. The deep impress which the Ishtar cult left upon
Semitic religious life has no parallel among the Aztecs.
Their culture was the product of agriculture and conquest
and not the culture of the oasis. Men had long been the
head of the family, and gods were at the head of their pan-
theon. The feminine element entered into their religious,
as into their social life under the conceptions of a polyg-
amous, and not of a polyandrous social order. The god-
desses may have been, and probably were, supreme in the
earlier days of the tribal life, but the conditions of the
country made the agricultural tribes an early prey to
other clans. Wars and conquests followed, producing
clans in which the virile elements of manhood were ideal-
ized, and in which gods soon became supreme. This
happened so soon that the earth goddesses never gained,
as in Arabia, where the environment made outside influ-
ences impossible and the deserts made the oasis type of
life predominant over everything, a character sufficiently
permanent to meet the shock of mixture and to survive
and absorb it.

In Peru a similar history can be traced. In the basin
of Lake Titicaca tribes known as the CoUa^ worshipped

1 Cf. History of the Conquest of Mexico, by William Prescott, Boston,
1858, Vol. I, p. 79 ; Frost's History of Mexico audits Wars, Vol. I, p. 42 ; and
Brantz Mayer's Mexico, Aztec, Spanish, and Bepublican, Vol. I, pp. 39, 40.

2 Cf. Payne's History of America, Vol. I, p. 324.

176 SEMITIC ORIGINS

as a mother, some of them the earth, and some the
lake,^ while the Yuncas or Yuncapata of the Pacific
coast thus regarded the ocean. ^ The dominant race
who produced Peruvian civilization, the Amyara and
Quichua, whose original home seems to have been in the
mountainous regions of what we call Bolivia and Argen-
tina, came thither and conquered the country. ^ Like
some other tribes,* the Incas were, in later times at least,
worshippers of the sun. They conquered the coast lands
also, and developed a high degree of civilization. Reli-
gious syncretism and a pantheon followed. Their social
order was definitely organized. Their lands were divided
into the lands of the sun, which supported the temple and
priesthood, the lands of the Inca, which supported the
king, and the lands of the people. The latter were divided
among them per capita.^ The priesthood was highly
organized and numbered about four thousand.^ The usual
features of agricultural social life present themselves in
their organization. The common people were monogamous,
and were not allowed to marry one from beyond the
bounds of their own community.'^ The nobles were polyg-
amous, while the Incas or sovereigns were extravagantly
polygamous. Honors practically divine were accorded
the Incas. One of the most striking features of the social
organization were Homes of the Selected, a kind of nun-
nery, where several hundred virgins were congregated,
and their chastity protected by the most stringent regula-
tions. These were destined to be the Inca's wives, or, if the
Inca chose, the wives of some of his nobles, or at times for
sacrifice to the sun. When he for any reason discarded
one after she had been destined for sacrifice, or after she

1 Cf. Payne, op. cit-, pp. 503-509.

2 Cf. Payne, ibid., pp. 376, 502 £f., also Vol. II, p. 555 fl.
8 Cf. Payne, op. cit., Vol. II, p. 560 ff.

* Cf. Payne, op. cit.. Vol. I, pp. 560, 561.

6 Cf . Presoott, Conquest of Peru, Vol. I, pp. 47-50.

' Presoott, ibid., p. 101.

' Presoott, ibid., p. 112.

TRANSFOKMATIONS IN BABYLONIA 177

had been taken to his palace and had lived with him for a
time, she returned to her native village.^ Like the Sem-
ites, they sacrificed only edible animals to their gods,^
regarded sacrifice as commensal, and concluded their
feasts with music and drinking .^ Like the Semites, too,
they had passed through a totemistic stage of develop-
ment before they reached the point of civilization which
has been described.*

At the time of the Spanish conquest, when Europeans
came in contact with the empire of the Incas, that race had
subjugated other tribes, and welded them into a complete
organization with its resulting pantheon. In this pan-
theon the sun god stood at the head, with his sister and
consort, the moon goddess.® Virachoca, a lake or rain
god, was also worshipped with his sister Choca.® These
deities were older than the sun deities, as their myths
show.'' They were survivals from the more primitive
social organization. Pochacoma, the animater of the
earth (a kind of Dionysos), also held a conspicuous place. ^
Cuycha, the god of the rainbow, and Chasca, a male
Venus, were also worshipped as attendants of the sun.*

In the religion of Peru, then, we find a course of develop-
ment quite similar to that of Mexico. The primitive god-
desses were retained, but in the religious syncretism of

1 Prescott, op. cit., Vol. I, pp. 109-112 ; E^Yille, op. cit., pp. 204-208 ;
and Payne, History of America, Vol. I, p. 564 ff.

2 Rgville, op. cit., p. 219.
» Prescott, ibid., p. 107.

* Cf. Prescott, ibid., p. 93 ; ESville, op. cit., p. 198 ; Sistoire des Tncas,
rois du Peru, par Jean Baudoin, Amsterdam, 1765, pp. 39 ff., 41 ffl. ; and
Payne's History of America, Vol. I, p. 445 ff.

6 Rgville, op. cit., pp. 153, 154 ; Prescott, op. cit., p. 93 ; and Baudoin,
op. cit., p. 80 ff.

« Note that the extreme practice of endogamy in Peru, similar to that
of the royal family of Egypt, had projected itself into their conception of
the gods, so that the celestials also married their sisters.

' Rgville, ibid., pp. 185-188.

8 Cf. Prescott, op. cit., p. 91, and Rfiville, op. cit., p. 189 ff.

» Prescott, ibid., p. 92, and Rgville, ibid., p. 194.

178 SEMITIC ORIGINS

the Inca's empire were subordinated even more than in
Mexico to male deities. This, as in Mexico, was no doubt
due in part to early conquests of the peoples who
worshipped the earth as a mother goddess, before that
worship became so fixed by long practice as to be able to
withstand, as the Semitic cultus of primitive times did, a
good proportion of this absorbing power.

If we turn to the gods and goddesses of ancient Greece, we
find similar beginnings and a similar result. The various
waves of races and of conquest which swept over Greece
finally left its pantheon a mixture, in which the male ele-
ment predominated. There are not wanting, however, evi-
dences that at the beginnings of its agricultural life many
goddesses of mother earth were worshipped. The most
obvious of these is Demeter, whose name probably meant
originally " Earth-mother " ; but, as Farnell has shown,
Artemis was such a goddess among the Greeks of Asia
Minor, ^ Hecate in jEgina,^ and Athena at one time at
Athens.* In the civilization which resulted from later
mixtures these goddesses lost their supremacy, with the
exception of Artemis, who in Asiatic cities like Ephesus
maintained her position, though under a somewhat trans-
formed character, till a comparatively late time.

There is some evidence that a similar history was
enacted in connection with the pantheon of Rome. Maia
seems to have been an earth-mother goddess, whose cult
was in a way maintained down to the latest times,* but
Rome became such a warlike power that the virile gods of
battle almost eclipsed in the historical period this primi-
tive goddess.

A similar history can in all probability be traced in the
Teutonic pantheon.^

1 See Farnell's Cults of the Greek States, "Vol. II, p. 464 ff.

2 Farnell, op. cit., Vol. II, p. 507.
8 Farnell, op. cit., Vol. I, p. 289.

* Maorobius, Saturnalia, I, 12.

* Cf. Gummere's Germanic Origins, New York, 1892, ch. xiv.

TRANSFORMATIONS IN BABYLONIA 179

From such examples as these the following conclusions
may be fairly drawn. Where the beginnings of agricul-
ture are possible men naturally worship goddesses which
they connect with the earth, or a lake, or some spring
which is conceived as the giver of fertility. ^ In all
probability as long as such communities remained peace-
ful such goddesses continued to be supreme, but when
other tribes or clans, attracted by their prosperity began
to conquer them, all soon became changed. These attack-
ing clans were in some cases pastoral, and consequently
patriarchal and worshippers of gods rather than god-
desses ; in other cases they were clans organized on the
republican basis for hazardous undertakings and therefore
worshippers of virility. If they had been agricultural
communities and worshippers of goddesses, warlike habits
in many cases changed these into gods. In the struggles
which foUowed the strong powers of the warrior would in
time become idealized by all as the chief powers of the
leading deity, and the old goddesses when retained would
take a subordinate position.

In an oasis country like Arabia the conditions were
somewhat different. The direct dependence of all upon
the oasis and the mother goddesses of these fertile spots
would keep even the republican clans, organized for the
caravan trade, largely dependent upon, and worshippers
of, the mother goddesses. The natural barriers of the
peninsula protected the clans from outside influences and
attacks, so that here even in the midst of long struggles
between clan and clan for supremacy the goddess could
maintain her position. We have a right therefore to ex-
pect that when the Semites went forth in hordes from
Arabia into other lands their mother goddesses would
present much more fixity of character than the mother
goddesses of ordinary agricultural communities such as
the Sumerians of Babylonia were. The peculiar emphasis

iThis is no doubt the origin of the " corn spirit " which Frazer traces
through so many countries. Cf. Golden Bough, Vol. II, passim.

180 SEMITIC ORIGINS

whicli the deification of the palm tree led them to place
upon sexual functions also gives a Semitic goddess a
character quite peculiarly her own. In the fragmentary
information which has come down to us we may not
always be able to distinguish these characteristics, but
where our information is full the task is not difficult.

In applying these principles to the gods of Babylonia
we are met by another difficulty. Were the Sumerian
communities always peaceful and their goddesses conse-
quently left in supremacy until the Semites invaded the
country and the struggle with them began ? The answer is
not easy, as it lies altogether in prehistoric time, but the
probabilities are all in favor of a negative answer. From
the mountains and high lands on either side of the Mesopo-
tamian valley pastoral or unsettled clans must have poured
themselves into the lower and more fertile lands of the
agricultural portions from the , time when the Sumerians
made their first settlements. At the very dawn of history
Eannadu and the kings of Kish had frequently fought and
conquered the Elamites,^ Eannadu boasting that he had
driven them back to their mountain, and we have no means
of knowing how many struggles between Babylonians and
Elamites may have preceded. In all probability such wars
had been going on for generations before the Semitic ad-
vent, and had as in Mexico, Peru, Greece, Rome, and else-
where transformed or subordinated the Sumerian goddesses.
If then we can find in Babylonia gods of tribes or cities
whose masculine character seems to have been fully estab-
lished before the dawn of history and whose traits seem to
have no organic connection with Ishtar or Athtar, we may
conclude that such deities are Sumerian. If, on the other
hand, we find cities where goddesses are supreme and where
the peculiar sexual features which were developed in Arabia

'Cf. Bevue d. assyr., Vol. IV, pi. 1, col. iii and col. v; also Radau,
Early Babylonian History, pp. 85 and 91 ; de Sarzeo's Dccouvertes, pi. 31,
No. 2, a and b, col. iii, and Radau, op. cit., p. 94 ; also OTBM., Ft. IX,
pi. 1 and 2 ; Hilprecht, OBI., No. 5, and Radan, op. cit., p. 128.

TRANSFORMATIONS IN BABYLONIA 181

are present, or goddesses connected witli the culture of the
date-palm, or gods developed out of goddesses which were
so connected, we may hold that the dominant element of
such civilization was Semitic.

In applying this test it will be most convenient to begin
with an old Babylonian kingdom, which has been brought
to light by the researches of recent years, the kingdom of
Kish. True, scholars have wavered as to whether it was
really a kingdom,^ but the fact that the name is followed
by the determinative for place when spoken of by those
outside its limits,^ seems to settle the matter. This city
or region appears to have been situated east of Babylon
and north of Shirpurla on the Tigris River in northern
Babylonia.^ It is the first of the Babylonian states whose
kings wrote their inscriptions in Semitic Babylonian, and
thereby reveal their Semitic origin.* One of the early
kings of Kish has left a votive inscription hitherto mis-
understood, which proves for the kingdom of Kish a
development of the Ishtar cult similar to that which we
have already proven by monumental evidence for south
Arabia.^ This inscription reads : —

1. " For the king of countries,

2. Nana, (Ishtar) ;

. 3. For the lady. Nana, (Ishtar),

4. Lugal-tarsi,

6. king of Kish,

6. the structure of a terrace

7. has made."®

iCf. WinoMer's AUorientalische Forschungen, 1st ser., p. 144, and
HUprecht, OBI., p. 270 (pt. 2, p. 56).

" Cf. Radau's Early Babylonian History, p. 126.

» Cf. Radau, op. cit., p. 112.

*Cf. Hilprecht, OBI., Nos. 5-10, and Scheil's Textes, elamites-
simitiques. Paris, 1900.

5 See above, Chapter IV.

« This inscription is published in GTBM., Pt. Ill, pi. 1, No. 12155. In
Sumerian it reads: (1) Dingir LXJGAL-RA KUR-KUR; (2) dingir

182 SEMITIC ORIGINS

In this inscription the name of the deity both in line 2,
where it is in apposition with " king," and is consequently
masculine, and in line 3, where the fact that it is preceded
by the word " lady " proves it to be feminine, is expressed
by the sign, which is employed as the ideogram for the
name of the goddess Ishtar,^ as the syllabaries and bi-
lingual hymns testify, but which scholars are accustomed
to read in Sumerian texts, "Nana" or "Ninnai."

It is clear from the argument in the first three chapters
of this work that the name Ishtar was no late invention of
the Semitic peoples, but had its origin in primitive Semitic
life. In all probability, therefore, it was carried by the
Semites with them to Babylonia, as well as to the other
countries whither they went. As the people of Kish were
Semitic, it was no doubt their name for the goddess,^ and

NANA; (3) NIN dingir NANA-RA ; (4) LUGAL-TAE-SI ; (5)
LUGAL KISH ; (6) GIR KISAL ; (7) MtJ-NA-RU. Thureau Dangin
translates (^Bev. W assyriologie, "Vol. IV, p. 74, n. 15, and Tablettes
chaldeennes inedites, p. 6, n. 15), "En I'honneur du dleu contr6es et
de Ishtar, de la dame Ishtar," etc. Radau (op. cit., p. 125, n. 3) would
render: "In honor of the god of countries and of Ishtar, the mistress
of the divine Innanna," etc. I believe both to have missed the signi-
ficance of the inscription as to the history of the development of
religious conceptions. Cf. my paper, " An Androgynous Baby-
lonian Divinity," in JAOS., Vol. XXI^, p. 185 £E. Eadau's reading
" Innanna " is based on Thureau Dangin's translation in Bev. semitique,
Vol. V, p. 67 ff., of Eannadu's Galet A, of col. ii, 1. 5. Thureau
Dangin translates the sign for Ishtar " Inanna " there because in col. v,
1. 26, he had been unable to render it othervrise. Upon reference to the
original publication in Bev. d'assyriologie, Vol. IV, pi. 1, it is found, as I
have pointed out in the paper referred to above, that it may equally well
be read in both passages "Ishtar" instead of "Inanna." Even if the
inscription was deposited by the king at a temple in Shirpurla, it vouches
for the conceptions prevalent in his own city.

1 Cf. II R., 59, 12, e.f. ; IV R., 1, 33 b ; and Briinnow's List, No. 3051.

2 The Sumerian name Nana, I take to be simply an epithet It is
usually written NA-NA-A. NA signifies "exalted" (cf. II R., 30, 24,
g, h, Briinnow, List, No. 1584, and Haupt, A8KT., p. 136, § 5). Words
are in Sumerian frequently repeated for emphasis ; thus we get NA-NA.
This repetition is too frequent in verbs to need illustration, but it also
occurs in nouns ; cf. UD-UD, OBL, No. 87, col. 1, 1. 46 ; and the final
A is the repetition of the final syllable in the emphatic state. Thus

TRANSFORMATIONS IN BABYLONIA 183

we have a right accordingly to read the sign Islitar in
both places where it occurs in the inscription of Lugaltarsi.
Clear evidence is thus presented of the development of
the old mother goddess into a masculine and feminine
deity at Kish, parallel to that of Athtar in Arabia, — a
development produced by the transformation in the social
structure caused by the changed environment.

If one is inclined to object to the conclusion just reached,
he might urge that this inscription very likely comes from
Telloh, and that accordingly it may not represent the
religious ideas of Kish at all, and that as Enlil is so con-
stantly called " King of countries " Ishtar may have been
written in the first line by a scribal error for Enlil. These
are two considerations which certainly deserve to be met.
To take the second one first, it may be remarked that in
the inscriptions from Telloh, Gudea frequently calls Nana-
Ishtar " Mistress of countries," ^ and that there was at
Shirpurla a god Lugal-Erim, who seems to have been but
another phase of Nana-Ishtar, " Mistress of Erim."^ Sup-
pose then that Lugaltarsi was addressing these, our con-
clusion would still be just, though it would apply to
Shirpurla instead of Kish.

But probably it applies to Kish, as the inscriptions from
Susa, recently published by Father Scheil,^ seem to confirm
it. The most important of these inscriptions is from

NA-NA-A means "the exalted one," and was applied as naturally to the
supreme mother goddess in primitiye times as sirtu (MAG.) was applied
to the same goddess, with a similar meaning in the Gilgamish epic
(of. Haupt Nimrodepns, p. 141, 1. 163. Jensen KB., Vol. VI, p. 241
renders it belit il&ni, "lady of the gods," which is an interpretation rather
than a translation).

1 Cf. e.g. Statue C, col. ii, 1. 2, and col. iv, 1. 10 ; i.e. de Sarzec's
Decouvertes, pi. 13, No. 1. Cf. Records of the Past, 2d ser.. Vol. II,
pp. 87, 89.

2 See de Sarzec, Decouvertes, pi. 8, col. ii, 1. 2, and col. iv, U. 8, 9.
Cf. also Records of the Past, 2d ser., Vol. I, pp. 75, 77, and KB., Vol. IIIi,
pp. 21, 23.

' Memoires de la delegation en Perse, Tom. II. Textes elamites-semi-
tiques, Paris, 1900.

184 SEMITIC ORIGINS

Manishtu-irba, king of Kish. It shows that the kings of
Kish had conquered Susa at a very early date. In other
archaic texts from Susa, written by men who were subject
to some foreign power, probably Kish, the ideogram for
Susa is the ideogram for Ishtar, plus ERIN, which means
"cedar forest."^ This same combination of signs is also
used to represent the name of a deity, which would accord-
ingly be " Ishtar of the cedar forest." ^ That deity is once
called "Lady" and "King" in the same inscription.^
Either, therefore, the Semites of Kish had planted the
worship of their goddess at Susa, where she became
metamorphosed into a god, or they had identified her, after
she had been so metamorphosed at Kish, with a god which
they found already at Susa. The result is for our purpose
the same in either case, for it confirms the development
suggested by the inscription of Lugaltarsi.

It will be best to examine next the gods of Shirpurla
(otherwise called Lagasli),* since we have more abundant
information concerning its pantheon than we have con-
cerning the gods of any other Babylonian city at a date
equally early. From the inscriptions of Gudea, who was
ruler of Shirpurla about 3000 B.C., it is possible to form a
tolerably clear idea of its principal deities for Gudea's
time, and the occasional glimpses which the inscriptions
of his predecessors give us of these deities, assure us that
substantially the same pantheon extended back to 4500
B.C., or earlier.^ The city or region of Shirpurla (for it

1 Scheil, op. cit., 58, 59, 63, 69.

2 Ibid., pp. 58, 59, 1. 8.

' Ibid., p. 69. It seems probable that this is the same deity which was
later called Khumbaba, and who dwelt in the cedar wood in the midst of
such magnificence. Cf. the Gilgamish epic, tablet V, Haupt, Nimrodepos
pp. 24 ff., 28, 54, 58, and KB., Vol. VI, p. 159 ff.

* Cf. Pinches, Ouide to the Kouyunjik Gallery, London, 1884, p. 7.

' See the inscriptions of tJrukagina and other early kings in KB., Vol.
III^, p. 10 ff., and in Radau, op. cit., p. 48 ff. Also the B inscription of
monument Blau, which must be much older. (Cf. Ward in PAOS., 1885,
p. Ivii, Jour, of Am. Arch., 1st ser.. Vol. IV, pi. v, and my translation,
JAOS., Vol. XXII, p. 123.)

TRANSFORMATIONS IN BABYLONIA 185

is not certain that it was simply one city)^ was, as Amiaud
pointed out^ and as other scholars have also observed,^
composed of four cities or districts, each of which possessed
its tutelary deity. These four districts were Girsu, in
which the god Ningirsu was the chief deity ; Uruazagga,
the chief deity of which was the goddess Ban ; Nina, over
which the goddess Nina presided, and a town the name of
which most scholars read as Gishgalla, but which I, with
Jensen, would read Erim, the principal divinity of which
was the goddess Ishtar or Nana. By the time of Gudea
these four places had long been united under one sov-
ereignty, and the four deities had been given places in one
mythological family. Ningirsu and Bau were husband
and wife, Nina was the sister of Ningirsu, while Nana was
perhaps his mother.* Shirpurla afforded at this time
many other deities beside these, but these were formed into
an amicable family, while most of the others were grouped
about them as subordinates. A few, like En-lil or Bel of
Nippur, were superior to these four. This superiority of
the gods of other cities had, however, grown out of previ-
ously existing political conditions, while the gods sub-
ordinate to this group had either been developed by the
application of epithets from a few primary deities, or bor-
rowed from other places.

If now we apply to the principal deities of Shir-
purla the rule formulated above,^ we reach the conclusion
that Uruazagga, Nina, and Erim were either Sumerian
settlements which had escaped war, or Semitic settlements,
because their chief divinities were goddesses. As the
former alternative is contrary to all probability, we are
driven to regard them as Semitic. This conclusion is con-

1 Cf. Ball, P8BA., Vol. XV, p. 51 ff.; Hommel, FSB A., Vol. XV,
p. 108 ff. ; and Davis, PA08., 1895, p. ccxiii ff.

2 Records of the Past, 2d ser., Vol. I, p. 46 ff.

3 See Davis in PAOS., 1895, p. ccxiii ff., and Price in AJSL., Vol. XVI,
p. 48 fl.

* Cf. Davis in PAOS., 1895, p. coxv.
' p. 179 fl.

186 SEMITIC OEIGINS

firmed by the traces of Semitic idiom which appear in the
inscriptions of Shirpurla.^ If this be true, these three
feminine divinities were three forms of the goddess Ishtar,
and it will be instructive to examine them a little more
closely.

To begin with Nana, it is clear from the preceding
argument that her real name was Ishtar, and that she was
probably so called in the popular speech of Erim. The
statements made concerning her by the kings would well
apply to Ishtar. Ur-Bau calls her " the brilliant, the
exalted lady," ^ and Gudea, " the bearer of the word of
life." ^ Davis has pointed out that Ninkharsag was
originally the same goddess as Ishtar, but worshipped
under a separate epithet, and the inscriptions bear out the
statement.* Under this epithet Entemena, about 4100
B.C., built a temple to her,^ Eannadu and Entemena claim
to have been nourished by her milk, as does Lugalzaggisi,^
and Gudea a millennium later calls her the " mother of the
city's children." ^ That Nana (Ishtar) was held in high
esteem in other ways is shown by the fact that Eannadu
claims that she gave him the patesiship of Shirpurla and

1 Cf. Kadau, op. cit, pp. 145-147. The language chiefly spoken in
Shirpurla was probably Semitic, but writing had been adopted here from
the Sumerians at a date long before the use of the cuneiform character
for the expression of thought in Semitic had begun in neighboring Baby-
lonian cities. It was apparently conformity to ancient custom which
maintained the use of Sumerian for the purpose of written expression so
long at Shirpurla, when at Kish, Guti, Lulubi, and Agade writing in
Semitic had been going on for some hundreds of years.

2 Cf . de Sarzec's Decouvertes en Chaldee, pi. 8, col. iv, 1. 9 ; cf. KB,
Vol. nil, p. 23.

' See Price's Great Cylinder Inscriptions A & B of Gudea, Cyl. A,
col. xiv, 1. 26.

* Cf. PAOS., 1895, p. ccxiv. ^ gee Radau, op. cit., p. 101.

^ Cf. Galet A, Sev. d'assyr.. Vol. IV, pi. i, col. ii, 1. 5, and my trans-
lation in J-O^^'., Vol. XXII, p. 186, n. 6. Cf. OBt, No. 115, and Radau,
op. cit., p. 118, and OBI., No. 87, col. i, 11. 28-29, and Eadau, op. cit.,
p. 133.

' Cf. de Sarzec, op. cit., pi. 20, col. i, 1. 3 ; Amiaud, Records of the Past,
2d ser.. Vol. II, p. 75, and Radau, op. cit., p. 198.

TRANSFORMATIONS IN BABYLONIA 187

the kingship of Kish.^ From such statements as these we
can, with the knowledge gained in the preceding pages,
fill out a tolerably correct picture of her character and
worship. She was simply the old Semitic mother goddess.
Since she was the tutelary deity of the city Erim, its
inhabitants were probably chiefly Semites.

There is also evidence that the Ishtar of the town
had, like the goddess worshipped by Lugaltarsi, begun to
undergo differentiation into a masculine and a feminine
deity. She is several times referred to as Lugal-Erim,
i.e. " king of Erim," showing that a confusion of thought
with reference to her sex had already begun.^

A goddess of this group about whom somewhat more is
known is Nina, the tutelary deity of the city or district
of the same name. She is represented in the inscriptions
by an ideogram, which is compounded of the ideogram
for house into which that for fish is inserted.^ This indi-
cates that she was previously the goddess of a fishing
town. The same ideogram was afterwards employed to
write the name of the city of Nineveh in Assyria. It
was, of course, used to express the name of the city of
Nina in Shirpurla. The name of the Assyrian city was
pronounced Nina or Ninua. As nun is the Semitic Baby-
lonian for fish, we have in the name Ninua a hint at what
men in the Assyrian period considered her name to mean.
Perhaps in Sumerian she had been called NIN-A, "lady
of waters," which was by a folk etymology afterwards
made in Semitic to mean "the fish." The reasons for
this we shall have occasion to examine by and by.

' Cf. Galet A, col. v, 1. 23 to col. vi, 1. 5. Text Mev. d'assyr., Vol.
IV, pi. i, and my translation JO AS., Vol. XXI i, p. 186, n. 6.

2 Cf. Bev. d'assyr., Vol. IV, pi. 1, col. ii, 1. 1.3 (cf. Thureau Dangin
in Mev. Sem., Vol. V, p. 67, and Radau, op. cit., p. 85), and de Sarzec,
op. cit., pi. 8, col. ii, 1, 2 (cf. Amiaud, op. cit., p. 75, and Jensen, op. cit.,
pp. 20, 21), also CTBM., Ft. X, No. 86900, U. 28, 29. The development
noted in the inscription of Lugaltarsi (above, p. 181 ff.) should be com-
pared.

8 Cf. Thureau Dangin's Becherches, No. 350.

188 SEMITIC ORIGINS

Our knowledge of the worship of Nina begins about
4300 B.C., with the inscriptions of Ur-Nina. He declares
that he built her temple, renewed her image, and caused
her servants to build for her two high places.^ The
word used for " servants " is expressed by the ideogram for
"dog," the Semitic term for sacred prostitute (see below,
Chapter VI). Eannadu and Entemena call themselves a
little later the " Chosen of her heart," ^ and Entemena built
various buildings for her.^ Ur-Bau calls her the mother
of the goddess Nin-mar,* and Gudea, the "mistress of
tablet writing,"^ "the child of Eridu,"® and "the supreme
lady."^ He says he built her temple and placed in it the
image of a lion.^ An inscription from the time of Dungi
calls that king " the lord whom Nina loves," and he calls
on an unnamed goddess, who was probably Nina, in behalf
of his life.®

From the general course of our argument we should
expect a goddess like Nina to be a form of Ishtar. .This

1 See de Sarzec, Decouvertes, pi. 2 in No. 2, KB., Vol. IIP, 11-15, and
Kadau, op. cit., pp. 61-63. Where Eadau reads "his wife for NinS," we
should read, " the lady NinS,."

2 Cf. Hev. d^assyr., Vol. IV, pi. i, col. i, 1. 9, and col. ii, 1. 1 ; also Bev.
semitique. Vol. V, p. 67 ; also de Sarzec, Decouvertes, pi. 43, and Radau,
op. cit., p. 116.

' Cf. Bev. cfassyr., Vol. II, pp. 148, 149, col. iv ; Radau, op. cit., p. 113.
Also de Sarzec, Decouvertes, pi. 31, No. 2, and Radau, op. cit., p. 94.

* De Sarzec, op. cit., pi. 8, col. v, 11. 8-10 ; cf. KB., Vol. IIP, p. 25.

6Cf. PSBA., Vol. XIII, pp. 62, 64, No. 2; Radau, op. cit., p. 193;
Gudea, Statue B in de Sarzeo's Decouvertes, pi. 16 ff., and KB., Vol. IIP,
p. 47.

« Cylinder A (cf. Price's Great Cylinder Inscriptions'), col. xx, 1. 16.

' PSBA., Vol. XIII, pp. 62, 64, No. 2.

8 Ibid.

9 The text is published in CTBM., Ft. V, No. 12218, and is as follows:
(1.1) Dingir (NIN) LIG (2) NIN-A-NI (Z) NAM-TI (4) Dingir DUN-GI
(4) NITAG LIG-GA (5) LUGAL URU-/u-MAMUG-KA (6) dingir
BA-U-NIN-A-AN (7) ZABAR ZID (8) UR dingir NIN-GIR-SU (9) EN
KI AKA dingir NINA-KA-KID (10) NAM-BA-KA-NI (11) MU-NA-
DIM, i.e. "To the powerful lady, his mistress for the life of Dungi, the
mighty hero, king of Ur, the exalted prince of Bau, lady of heaven, the
brilliant, the faithful one, servant of Ningirsu, the lord who is beloved of

TRANSFORMATIONS IN BABYLONIA 189

expectation is confirmed by the fact that she had a com-
pany of prostitutes, and was probably the lady of life. It
is further confirmed by the fact that Entemena declares in
several different inscriptions that he built a storehouse of
dates for Nina.^ The date-palm we have seen to be so
closely connected with the Semitic goddess that this
becomes another evidence that Nina was a form of that
divinity. Gudea, as we have noted, says that he placed
in Nina's temple a statue of a lion. Dr. Ward has called
attention to a seal in the Metropolitan Museum in New
York, which represents a nude goddess riding on a lion
drawing a chariot and holding the lightnings in her
hands.2 He dates the cylinder at 3500 to 4000 B.C. It
seems to me probable that this is a representation of Nina,
and that it gives pictorial evidence of her close relation-
ship to Ishtar.

The worship of the third of these goddesses, Bau, is
known to us from the same early sources as that of Nina,
since she is mentioned in the inscriptions of Urkagina,^
Ur-Nina,* and Gudea. ^ She is easily shown to be a form
of Ishtar. Ur-Bau calls her the "good lady,"^ while
Gudea calls her his chief mistress,^ and has left on record
two prayers in which he applies the term mistress in

Nina, the beauty of her .building constructed." Of course the goddess
addressed in the first line may be Ishtar or Bau, but since the prince calls
himself the beloved of NinS, it seems more probable that he addresses her
throughout.

1 See e. g. OTBM., Pt. X, No. 86900, 11. 14, 15, and de Sarzec's Decou-
vertes, PI. 5 bis, Face, col. iv, 11. 2, 3.

2 Cf. AJSL., Vol. XIV, p. 95 ; cf. PSBA., Vol. XVIII, pp. 156,
157.

8 Cf. Amiaud in Becords of the Past, 2d ser.. Vol. I, p. 69, and Radau,

op. cit., p. 50.

* Cf. Bev. d'assyriologie, Vol. IV, p. 106, No. 11, and Radau, op. cit,

p. 65.

6 See e. g. KB., Vol. IIV-, pp. 58, 59.

8 Cf. de Sarzec Decouvertes, pi. 8, col. iv, 1. 3 ff., Amiaud, op. cit., p.
76, and KB., Vol. Ill, p. 23.

' Cyl. A, col. xxiv, 1. 6.

190 SEMITIC ORIGINS

several different ways and prays especially for life,^ as
though she were the life giver. In nnother passage he calls
her "mother Bau."^ Galalama calls her also " mother (?)
of Shirpurla."^ On New Year's day, probably at the
beginning of the first month, Gudea tells us also that he
celebrated the festival of the goddess Bau, offering her
various sacrifices of oxen, sheep, and lambs ; dates and
shoots of palm forming also a prominent feature of the
offering.* This spring festival was, as we have shown in
a previous chapter, the old spring festival of the yeaning
time, and was a festival of the goddess Ishtar. A mother
goddess, whose festival celebrated the birth of young,
would, among a Semitic people, be a form of Ishtar. The
name BA-U was simply an epithet, meaning "producer of
food," and was probably given her as the goddess of the
date tree and then of agriculture in general. Bau had a
brother, Ningishzida,^ whose name means "lady (or lord
(?)) of the tree of life." Jastrow takes him to be identical
with Ningirsu,^ but Price considers this to be impossible.''
In view of the development at Kish noted above,^ where
Ishtar was divided into a masculine and a feminine deity,
it is probable that Ningishzida was originally an epithet
given to Bau in consequence of her connection with the
palm tree, and gradually, as she was differentiated, con-
tinued to be applied to the masculine portion of her.
Possibly it may seem more probable to some that Ningish-
zida was a name of Tammuz, and that he is, in consequence
of this, a brother of Bau-Ishtar ; the result is in this case

1 Cf. de Sarzeo, op. cit., pi. 13, Nos. 1 and 4, Amiaud, Becords of the
Past, 2d ser., Vol. II, pp. 92, 103, and Radau, op. cit., pp. 202,
208.

2 Cyl. B, col. xvli, 1. 2.

= Cf. de Sarzeo, op. cit., pi. 21, No. 4, and KB., Vol. IIP, p. 71.

* Cf. KB., Vol. Ill, p. 61, and Amiaud, op. cit., p. 101.

6 Cf. Cyl. B, col. xxiii, 1. 5, and Davis, PAOS., 1895, p. coxv.

* Beligion of Babylonia and Assyria, p. 92.
' AJSL., Vol. XVII, p. 60.

8 p. 181 ff.

TRANSFORMATIONS IN BABYLONIA 191

the same, though the goal by which it is reached be
slightly different.^

At the time from which most of our information comes,
the four districts of Shirpurla had long been united under
one sovereignty. This had produced religious syncretism,
and the gods were formed into a pantheon. In this pan-
theon Bau was regarded as the wife of Ningirsu, the god
of Girsu, the fourth of the districts of Shirpurla.^ She is
also said to be the daughter of Anu,^ but this, as we shall
see by and by, has less historical significance than some of
the mythological statements concerning other forms of the
goddess.

The view we have taken of the nature of Bau is con-
firmed in another way. One of her titles was Gatumdug.*
Indeed, under this name she became almost a separate god-
dess. As Gatumdug she is frequently called "mother of
Shirpurla,"^ but is shown to be originally identical with
Bau since she is said to sit enthroned in Uruazagga,^ Ban's
city.

Ningirsu, the god of Girsu, the fourth district of Shir-
purla, is mentioned in the inscriptions much oftener than
any of the goddesses of the other districts. Most of the
references, important as they are for a knowledge of other
phases of the religion, do not materially help us in solving
the problem of origins.'' The application of our econom-

1 Other epithets of Bau were Ma-ma (of. OBI., Pt. II, p. 48, n. 6),
and Chi-la. In later times, this last name prevailed (cf. Jastrow, Eeligion
of Babylonia and Assyria, pp. 60, 105, 166, etc.), and became the goddess
of healing and of the nether world.

" Cf. Davis, op. cit., p. ccxiv.

» Cf. KB., Vol. IIP, p. 53, Amiaud, op. cit.. Vol. II, p. 91.

* Cf. Amiaud, op. cit.. Vol. I, p. 60.

' See, e.g., de Sarzec, op. cit., pi. 14, col. i, 1. 1 fi., Amiaud, op. cit.,
Vol. II, p. 97.

• Cf. de Sarzec, op. cit., pi. 14, col. iii, 1. 6 ff., and Amiaud, op. cit., Vol.
n, p. 99. Gatumdug is once an epithet of Nana, see below, p. 260, n. 6.

' He is called In them, " the king," "the great warrior of Enlil," etc.,
epithets which are important as showing his position in the pantheon of
Shirpurla, but which throw little light on his original nature.

192 SEMITIC ORIGINS

ico-religious test has made it clear that the towns or dis-
tricts of Erim, Nin&, and Uruazagga were peopled mainly
by Semites, and probably founded by Semites. At all
events, at the very dawn of history, the Semitic element
in the civilization is predominant. Is the same true of
the remaining district of Shirpurla, Girsu ? I think that
it is, for the following reasons : 1. The name Ningirsu
really means "lady of Girsu." ^ It is true that in at least
one phrase the Sumerian NIN seems to mean " lord," ^ but
it has almost universally a feminine signification,^ which
was no doubt its primary meaning. The Sumerians had
another word for "lord" (viz. EN), and they can hardly
be supposed to be so lacking in the sense of sex as to have
expressed at first both "lord" and "lady" by the same
word. Confusion between the two words would be very
natural, however, if a goddess had been metamorphosed
into a god as was done in South Arabia and at Kish. The
word in the name which once had a feminine meaning
would then seem to have a masculine signification, thus
producing the confusion. 2. At the time of Gudea, both
masculine and feminine qualities may in one passage still
be traced in the conception of this god. Certain gifts are
presented in one line to " mother Ningirsu," and two lines

1 The name of the city or district is usually spelled GIR-SU, but in
two or three inscriptions in de Sarzec's Decouvertes (of. No. 4 bis., and
KS., Vol. nil, p, 10, col. i, 1. 5, and Radau, op. cit., p. 58, n. 6), it is
spelled SU-GIR, or SUN-GIR (of. Hommel, Sum. Les., No. 7). Some
scholars have therefore taken it to be identical with the later Sumir, cf.
Radau, ibid., and Rogers, History of Babylonia and Assyria, Vol. I, p.
356, who reads it Sungir. If GIR-SU be the spelling, it probably meant
" body lance," perhaps equivalent to " girdle lance." This was at least
an early folk interpretation of the name, for in a Babylonian inscrip-
tion from before 5000 e.g., the Blau Monuments, a certain Khakhatabbar
says that Ningirsu's monument of protection, a lance (GIR) he brought
and placed in his temple. The inscription is inscribed on an object shaped
like a lance blade (cf. American Journal of Archaeology, 1st ser., Vol.
IV, pi. V, and my article "Notes on the Blau Monuments" in JAOS.,
Vol. XXIII, p. i2,S),

-! AL.,^ No. 301% ^i,4 No. .309''.

" Cf. Brunnow's List, Nos. 10984-10990.

TRANSFORMATIONS IN BABYLONIA 193

farther on, to "lord Ningirsu.''^ This confusion can onlj-
be accounted for by supposing that there was a time when
Ningirsu was the goddess of Girsu, and that she had after-
wards been transformed into a god. 3. A bilingual frag-
ment of a later time equates EN-GIR-SI (the word
NIN, "lady," having been here changed to EN, "lord"),
with Tammuz.2 This indicates that Ningirsu and Tam-
muz were closely related, just as Ishtar and Tammuz were.
If Ningirsu were a transformed Ishtar, the Tammuz which
originally accompanied her may well have been fused with
the resultant god either during or after the process of
transformation. 4. If Ningirsu had been a Sumerian
deity, the prehistoric wars would probably have completely
effected his transformation from a goddess before the
dawn of history. 5. Ningirsu came to mean husband-
man,^ — a meaning which appears to have been derived
from the fact that he was a god of fertility and life, such
as a transformed Ishtar would be. 6. Entemena, in a
mutilated text on an old gate socket, the reading of which
is not quite certain, appears to call him " god of life," * as
the goddesses of Shirpurla are called "lady of life."
7. The scores of phallic shaped cones inscribed to Nin-
girsu, found at Shirpurla, such as are pictured in de Sar-
zec's JD&couvertes en ChaldSe, pi. 38, point to a connection
of his cult with the sexual cult of Ishtar. For these

1 Cyl. B, col. X, U. 5, 7. The passage reads (1. 3) GISTIN-A DA
GISTIN-TIN-A DA (4) UZ AZAG UZ GA GU BIR-MIR (5) AMA
dingir NIN-GIR-SU-KA (6) NI-GA-BI ES E-SI-A-MUS NU-GUB-
DA (7) EN IMIR SIBA BIR-IMIR EN dingir NIN-GIR-SU-RA
(8) MI-NI-DA MU-NA-DA DIB SUM, i.e. "Wine lie brought up,
strong drink he brought up, a goat, a perfect goat, milk, the drink from
the asses of mother Ningirsu, the cream of their milk in the temple Eshia-
mush he offered ; to the lord of the asses, the shepherd of the asses, the
lord Ningirsu he raised, he lifted up, he brought, he presented it."

2 IV R., 27, No. 6, col. ii, 11. 42-43.

8 Cf. V. R., 16, 39 ef, and Brunnow's List, No. 10995.

< CTBM., Pt. V, No. 12061, 11. 10-12. They read dingir NIN-GIR-SU

LUGAL(?) DINGIR-A-NI DINGIR TI(?), i.e. "Ningirsu

the king, his god, the god of life." The sign TI, "life," is not quite
certain, as it is partly erased.
o

194 SEMITIC ORIGINS

reasons it seems highly probable that Ningirsu was a mascu-
linized Ishtar. The fact that he was transformed, while
Nana, Nina, and Ban were not, is probably due either to
the fact that Girsu was the conquering and the more war-
like of the settlements of Shirpurla, or to the presence of a
larger Sumerian element there. It seems probable that
Girsu was the oldest of the four settlements. Ningirsu,
its god, appears already before 5000 B.C., under that name
on the Blau Monuments.^ This presupposes the existence
of the city at that early time. There is, however, in the
inscription nothing to indicate whether Ningirsu was at
that time masculine or feminine (unless the fact that the
sacrifice consisted of ewes may point to a goddess), as the
suffix used in referring to the deity may be used in either
gender.2 Girsu, too, was the original seat of the monarchy,
which afterwards conquered the other districts, for Ur-
kagina, about 4500 B.C., styles himself indifferently king
of Girsu or king of Shirpurla. Girsu must have been
originally quite separate from the other districts. Indeed,
each was no doubt originally quite an independent settle-
ment. That that settlement was predominantly Semitic
is shown by the Blau Monuments, for the superior race
who are pictured upon them have the Semitic nose, while
the inferior or slave race which they show, have quite
different features.^

The conclusions here reached are not at all in conflict
with the view that in the period from which our inscrip-
tions come Ningirsu was a sun god ; they simply show
that before he was identified with the solar orb he was a
chthonic mother goddess.

No doubt some Sumerian elements beside their sys-
tem of writing entered into the Semitic civilization of

1 Cf. my article "Notes on the Blau Monuments " in J'^OiS'.jVol. XXIF,
p. 123.

2 It is NI ; cf. Brtinnow's List, Nos. 5330, 5331.

' Cf. American Journal of Archaeology, 1st ser., Vol. IV, pis. iv
and V, and Ward, ibid., p. 40.

TEANSFORMATIONS IN BABYLONIA 195

Shirpurla, but it is impossible at this distance to tell how-
great they were. Possibly some of the elements which
helped to transform Ningirsu into a god were Sumerian.
It is not necessary from the religious point of view to
postulate any large amount of such influence ; the laws of
Semitic social evolution are in the Mesopotamian environ-
ment sufficient to account for all that occurred in the
realm of the religion.

Before passing from Shirpurla to other parts of Babylonia
it will be convenient to remind ourselves that we are seek-
ing origins which lie altogether beyond the horizon of
written history and which can only be reconstructed by
working backward from sporadic survivals. In this re-
construction help may often be obtained from the myths
which grew up around the pantheon of Shirpurla and
other cities. Even at the early date when the written
history of Babylonia begins, the country had been united
in various political organizations till some gods like Enlil
and Enki (Ea) had become largely dissociated from their
original habitations and had entered into various pantheons
as lord of the earth and of the deep. Anu is a god whose
local habitation 1 we cannot trace, and who seems in the
historical period to have been more of an abstraction than
Enlil and Enki He had been added to these, and the
three had been formed into a triad. In this triad Anu in
theory stood at the head, but in practice the other two
were more honored. There are, therefore, three classes of
myths to be distinguished at the very dawn of Babylonian
history ; (1) those which recall migrations of tribes or
parts of tribes from earlier places of residence, like the
myth that Nina is the daughter of Enki or Ea ; ^ (2) those
which resulted from a long political subjugation, like the

1 See Jastrow's History of Babylonia and Assyria, p. 89. The state-
ment of Jeremias (Cliantepie de la Saussaye's Beligionsgeschichte, "Vol. I,
p. 171), that he was the god of Ereoh, is a misconception. One might
with as much reason call him a god of Shirpurla. See below, p. 218 ff.

"IV E. 1, col. ii, 1. 38 ; cf. Davis, PAOS., 1895, p. coxv.

196 SEMITIC OEIGmS

myth that Ningirsu is the son and warrior of Enlil ; ^ and
(3) those which have grown up out of the later abstract
conceptions, or the identification of the gods with celestial
objects, like the myth that Bau is the daughter of Anu.^
Careful examination will often enable us to distinguish
these different classes of myths from one another, and to
do so will aid us in following our slender thread of evi-
dence through the tangled mazes of Babylonian life.

Returning now to the first of the three myths just men-
tioned, we are led to the city of Eridu, the most southerly
of the old Babylonian towns, and to its god Ea. Nina is
called the daughter of Ea and the child of Eridu. ^ Eridu
was the city of the god Ea, its ideogram being the same
as that of the god with the determinative for place affixed.*
It was, about 4000 or 5000 B.C. or earlier, situated on the
shore of the Persian Gulf.^ The name of Nina was, as we-
have seen, written with the sign for house around the sign
for fish ; while Ea was often pictured under the form of a
fish, or as clad in a fishskin. A legend preserved for us
through Berossos and Eusebius tells us how Oannes (who
is certainly identified with Ea^) bore the form of a fish,
and how he came up by day to the land and taught men
how to construct houses, till the earth, collect fruits,
compile laws, and all other useful knowledge.'^ In the
pictorial representations Ea is seen as part man and part
fish.^ The fact that the fish form enters into the repre-

1 See e.g., Cyl. A, col. vii, 1. 5, col. viii, 1. 21, and Cyl. B, col. vi, 1. 6 ;
of. Price, AJSL., Vol. XVII, p. 49.

2 See above, p. 191, and below.
^See above, p. 188.

* Cf. Brunnow's List, Nos. 2625, 2645, and 2649.

5 Cf. Sayce, Hibbert Lectures, p. 104, and Peters, Nippur, Vol. II,
p. 299.

^ Cf . Lenormant, Histoire ancienne de Vorient, Vol. V, p. 231 S.,
Sayce, Hibbert Lectures, p. 1.31, and Peters, Nippur, Vol. II, p. 299.

'Cf. Cory's Ancient Fragments,^. 23 ff., also a cuneiform original
of a part of it, published by Soheil in Becueil de traveaux. Vol. XX,
p. 126 ff.

8 Cf. Lenormant, op. cit., pp. 232, 238, and Sayce, op. cit., p. 133.

TRANSFORMATIONS IN BABYLONIA 197

sentations of both Nina and Ea confirms the statement of
the mythology that the two were kindred. Sometimes
the fish god is pictorially represented as applying the fer-
tilizing cone to the sacred palm tree,^ and this gives an-
other thread of connection between the two, for as noted
above, Nina was the goddess to whom dates were sacred. ^
At Eridu, the city of Ea, there was also a sacred tree,^
no doubt a palm, so that it is no accident that the fish god,
or god of the water, is represented as fertilizing the palm.
An unpublished cylinder in the British Museum repre-
sents Ea thus and calls him "the god of life."* George
Smith called attention to the fact that an unpublished
brick in the same museum is inscribed to Ea under the
name Nin-Eridu, or "lady of Eridu. "^ Amiaud doubted
whether it really applied to that god,® but there can, in
view of the development which we have traced elsewhere,^
be no doubt of it.

The meaning of all these facts and myths would seem to
be this : Eridu was probably the oldest Semitic settlement
in Babylonia. Hither from Arabia the Semites came and
planted their earliest colony, probably selecting the site
because they found their sacred palm tree already growing
there. The proximity of the Persian Gulf led them in
course of time to associate their goddess with that body of
water as they had in Arabia associated her with the spring
or well of the oasis. If a Sumerian fishing goddess pre-
ceded her here, identification of the two may have hastened
the process. That in time she should be associated with
the fish symbol was perfectly natural. In the lapse of
years colonies were sent out to other points, partly in
consequence of the natural multiplication of the populace,
and partly in consequence of new immigration from

1 See Lenormant, op. cit. , p. 232. ^ p i89_

«Cf. Jensen, Kosmologie, p. 99 ff., 249, n. ; Hilprecht, OBI., pt. 1,
p. 28, and Radau, Early Babylonian History, p. 231.

« Sayoe, op. cit., p. 133. ^ tSBA., Vol. I, p. 32.

• Becords of the Past, 2d ser.. Vol. I, p. 60.

' E.g., that in Athtar, Ishtar at Kish, and Ningirsu.

198 SEMITIC ORIGINS

Arabia. Nina was one of these, but probably by no
means the earliest. The first colonists at Eridu brought
with them the artificial culture of the palm tree, a culture
probably before unknown to the country, and this addi-
tion to such knowledge of agricultural pursuits as the
country may have possessed before led in process of time
to the myth that Ea was the source whence all knowledge
of agriculture and civilized pursuits came. As this myth
grew Ea became in consequence the god of wisdom. He
was so regarded by Eannadu,^ Entemena,^ Lugalzaggisi,'
and through all the subsequent history. As Semitic
society was in this environment organized on another
basis, the Ishtar of Eridu was transformed into a god,
as happened also at Kish and Girsu. As early as the
time of Eannadu the transformation had occurred, since
he is called " king " * by that monarch as he is centuries
later by Ur-Bau^ and Dungi.^

The worship of Ea is so widespread in Babylonia at the
first dawn of history that we are compelled to suppose
that in prehistoric time Eridu had been the seat of an
empire which held sway over all of southern Babylonia.
There is no improbability to offset this necessary inference
from the phenomena of the religion, but on the other hand
as the oldest Semitic settlement it would be very natural
for it to become the head of a kingdom.

After the time of Khammurabi (about 2300 B.C.) Ea
was regarded as having for a spouse the goddess Damkina,
" the lady of the earth." She does not appear in the older

1 Cf. Bev. d" assyriologie, Vol. IV, pi. i, col. ii, 11. 6, 7 ; Rev. semi-
tique, Vol. V, p. 67 ; and Radau, Early Babylonian History, p. 84.

2 Sev. d'assijriologie, Vol. IV, pi. ii, col. v, 11. 24, 25 ; cf. Thureau
Dangin's translation, ibid., p. 49, and Radau's op. cit., p. 108.

" See Hilprecht, OBI., No. 87, col. i, 11. 17, 18 ; cf. Radau's translation,
op. cit., p. 132.

* Cf. Radau, op. cit., p. 80.

6 Cf. de Sarzec, Decouvertes, pi. 8, col. iv, 1. 11 ; cf. KB., Vol. IIP,
p. 23.

s See Winokler's Altorientalische Forschungen, 1st ser., p. 547, No. 8 ;
and Radau, op. cit. , p. 224.

TRANSFORMATIONS IN BABYLONIA 199

literature,^ and is clearly simply a female outgrowth or
counterpart of the Ea who two thousand years earlier had
become a male deity, and who had as the necessities of the
pantheons required, taken on more specialized functions.
His own reflection was at last assigned to him in female
form for a spouse, that he might not stand alone.

Returning to the pantheon of Shirpurla and following
another mythological clew, we are led by the statement of
Arad-Sin, king of Larsa,^ that Ishtar (Nana) of Khallabi,
which was, perhaps, a colony of Erim,^ is the daughter of
EN-ZU or Sin, to the city of Ur, of which Sin was the
chief deity. In the older texts the ideograms for Ur and
Sin are as identical as those for Eridu and Ea. Next to
Eridu, Ur was the most southerly of the ancient cities
of Babylonia. We do hear in mythological poetry of
Surippak, farther to the south, which was buried in the
flood, but it plays no part in the history. Ur was a little
to the westward of Eridu,* and is represented by the mod-
ern mound of Mugheir. It was a very old city, probably
not appreciably younger than Eridu. Its kings at various
times held sway over the rest of Babylonia, and probably
had done so before the dawn of history, for we find Sin,
the god of Ur, a member in high standing of the pantheons
of other cities when first the written records give us
glimpses of their life. Naram-Sin, king of Agade about
3750 B.C., is said by his name to be the favorite of this
god of Ur, and gives us other evidence that Sin was wor-
shipped beyond the borders of Ur.^ Lasirab^ of Guti

1 Cf. Jastrow's Beligion of Babylonia and Assyria, p. 64.

" PSBA., Vol. XIII, pp. 158, 159; Davis, FAOS., 1895, p. ooxvi,
evidently quoted from memory when he ascribed this statement to
Gudea.

' See below.

* Cf. Peters, Nippur, Vol. II, p. 296 ff., and Rogers's History of Baby-
lonia and Assyria, Vol. I, p. 290.

* Cf. Thureau Dangin, Oomptes rendus de Vacad. inscr., 1899, p. 348,
and Radau, op. cit., p. 17.3 ff.

6 Cf. Winckler, ZA., Vol. IV, p. 406, and Radau, op. cit., p. 175 ff.

200 SEMITIC ORIGINS

and Anu-banani^ of Lulubi had both worshipped him at
a still earlier time. The worship of this god also appears
at Shirpurla, where he was probably regarded as the father
of Nana or Ishtar.^ This fact, like the myth that Nina
was the daughter of Ea, would seem to give a hint that
Erim, the district over which Nana ruled may have been
colonized from Ur, as Nina was colonized from Eridu, or at
least that there was some close connection between them.
If this be true, we must suppose that Sin was at the first
an Ishtar and was transformed into a male like Ishtars in
other places. This view cannot be made out as clearly
as the cases which have already been treated, but its
probability is increased by the following considerations :

1. Dungi calls the chief deity of Ur Nin-Ur, "lady of Ur,"
and his "mistress," as Ea was called "lady of Eridu."*

2. An old hymn to the moon god* attributes to him the
authorship of all fertility in a way quite explicable if he
had first been the chthonic mother goddess, but which
would be meaningless were he simply a personification of
the moon. He is in this hymn called " lord of increase,"
" the begetter of everything," " the begetter of gods and
men," the " maintainer of the life of the world," the one
"at whose command vegetation is created,"^ etc. These
are all epithets not only befitting Ishtar, but which in
another old hymn are most of them actually applied to
her.^ 3. Throughout the hymn and in other texts Sin is
constantly called " father," ^ which would be very natural
if he had grown out of a mother goddess. 4. His symbol

1 Cf. Becueil de traveaux, Vol. XIV, pp. 100-106, since published in
Textes elamites-semitiqites, andBadau, op. cit., p. 177.

2 Cf. Amiaud in Becords of the Past, 2d ser., Vol. I, p. 57, and above
p. 185.

3 Cf. Hilprecht, OBI., No. 16, and Radau, op. cit., p. 224.
<IVR., 9.

' See Jastrow, Beligion of Babylonia and Assyria, pp. 303, 304.
' Cf. Haupt, A8KT., p. 116 ff., Zimmern, Babylonische Busspsalmen,
p. 33 ff., and Hebraica, Vol. X, p. 15.

' See, e.g., the text of Nabu-na'id, I R., 69, and KB., Vol. Ill 2, p. 81 ft.

TRANSFORMATIONS IN BABYLONIA 201

was the ox. He is called in the hymn (11.19, 20) "the
strong bull with great horns." The symbol of Ishtar, as
shown on an old seal, was the cow,^ and such a transforma-
tion as we have supposed would connect the two. Simi-
larly the symbol of Ashtart at Tyre was a cow, that of
Baal and Yahwe in Canaan, and of Athtar in South
Arabia, a buU.^ We cannot, however, press this con-
sideration, since the bull and cow are found as divine
symbols in many agricultural communities where there is
no possibility that an Ishtar or an analogous god had pre-
ceded. There can be no doubt though that a large Semitic
element entered into the make-up of the moon god of Ur,
but the possibility that that Semitic element took on the
bull symbol through Sumerian influence must be recog-
nized. The fact that a similar conception prevailed on
purely Semitic soil in southern Arabia renders, however,
the supposition of Sumerian influence unnecessary. 5. The
ideogram by which Sin is represented in many Sumerian
texts, EN-ZU, means " lord of knowledge, or of might, or
of wisdom, or of increase," ^ any or all of which are mean-
ings which would naturally spring from the conceptions
entertained of Ishtar.

Perhaps we shall get more light on this matter when
Mugheir is excavated and its earliest texts recovered, but
we are at present justified in regarding Ur as a city pre-
dominantly if not altogether Semitic, and its god as in
large degree the result of the absorption of an Ishtar,!or more
probably simply a transformed Semitic goddess. If this be
the real origin of Sin, the development must have been
completed very early, for a very archaic text* dating prob-
ably from before 5500 B.C. seems to call Sin Ab, "father."

1 See Scheil, Eecueil de traveaux, Vol. XX, p. 62.

2Cf. Hebraica, Vol. X, p. 31, ZDMG., Vol. XXX, p. 289 (of. Se-
braica, Vol. X, p. 56), and the articles "Bull" and "Calf, Golden," in
Hastings's Dictionary of the Bible, Encyc. Bib., and Jewish Encyc.

8 Cf. Brunnow's List, Nos. 130-137.

* Cf. below, p. 213, n. 5. The interpretation of the tablet is as yet
tentative.

202 SEMITIC ORIGINS

After the deity of Ur had been identified with the
moon, the hymns and prayers addressed to him are largely
occupied with praises of his brightness and other quali-
ties which were suggested by the brilliance and the
movements of the moon.^ In the pantheons of the his-
toric period Sin, like other gods, was called the son of
Enlil.2 This resulted probably from a long prehistoric
hegemony on the part of Nippur, Enlil's city, of which
more will be said below. Either the result of this hege-
mony, or the fact that Sin was a younger deity, prevented
him from ever occupying the same exalted position as
Enlil, or even as Ea. He was later a member of the
second triad in the pantheons of Babylonia, but not like
the others of the first.^

Taking once more as a point of departure a mythologi-
cal statement from the pantheon of Shirpurla, we are led
by the myth that Ningirsu was the son and warrior of Enlil
to consider next the god of Nippur.* In antiquity his
shrine at Nippur rivals the shrine of any other Babylonian
god.^ Among the earliest inscriptions yet published, except
such as the Blau Monuments, are some by a certain Enshag-
kushanna, lord of Kengi or Sumir, who before 4500 B.C.
devoted to Enlil, " king of countries," the spoil of his vic-
tories over Kish.^ This shows that even at that remote
period Enlil had come to be regarded far outside the con-
fines of Nippur as the principal god, and from this time

1 See, e.g.. King's Babylonian Magic and Sorcery, p. 5.

2 Cf. Ur-Gur, I R., 1, No. 5, KB., Vol. IIIi, p. 79.

' The inscriptions of the kings of Ur throw little light on the character
of Sin beyond the fact that he was regarded as a king (cf. KB., Vol. Ill i,
pp. 77, 93). The moon god was also worshipped at Harran, hut of the
origins of his worship there we as yet know nothing. Probably much
the same history could be written of his worship there as that which we
have sketched for the moon god of Ur, had we the material.

* See above, p. 195 ff.

6 Cf. Hilprecht, OBI, Vol. I, Ft. 2, pp. 44-46, and Peters, Nippur, Vol.
II, p. 246.

6Cf. Hilprecht, OBI., Nos. 90-92; Radau, op. cit., pp. 44, 45; and
Rogers, History of Babylonia and Assyria, Vol. I, p. 351 ff.

TKANSFORMATIONS IN BABYLONIA 203

onward lie was honored by all worshippers, of whatever
city, more than any other god.^ Even the rulers of Shir-
purla made their own god Ningirsu subject to him. This
fact can, I think, be adequately accounted for only on the
supposition that Nippur had been in prehistoric time the
head of a kingdom which included all of Babylonia.

At a time almost as early, — a time before the Sargonic
period, — Enlil had a female spouse, Ninlil. Urzaguddu,
a king of Kish, ^ and Anu-banini, of Lulubi,^ both wor-
shipped this pair, and in later times they are often grouped
together.

In all probability there is in the Enlil of Nippur a large
Sumerian element. The worship which he received from
men of all cities is no doubt to be accounted for in part by
the political supremacy of Nippur, as already suggested,
but in part too by the fact that he was an old pre-Semitic
god of the soil. Semites, when first they went into a new
country, thought it necessary as late as the eighth century
B.C. to learn the worship of the god of the land* in order
to reside there safely, and in the earlier times they would
have this feeling in still larger degree. The Semites in
coming into Babylonia would therefore adopt, in some
measure, the worship of its native gods wherever they
settled, while they kept also the worship of their own
goddess. If this foreign worship were practically unor-
ganized, it would make little impression, and would leave
the worship of the mother goddess comparatively pure ;
but if it had assumed a definite form, it would, in fusing

1 Cf. Jastrow, Meligion of Babylonia and Assyria, p. 52 ff.

2 Cf. Hilprecht, OBI., No. 93, and Eadau, op. cit., p. 125, n. 1.

" Cf. Becueil de traveaux, Vol. XIV, pp. 100-106, and Radau, op. cit.,
p. 177.

* See 2 Kgs. n^-^i; cf. also Budde, Beligion of Israel to the Exile,
pp. 53-55. An instance of fusion, not altogether dissimilar, from a much
later time is found in an Aramaic inscription of the second centuiy b.c.
from Kappadocia, published in Lidzbarski's Ephemeris far semitische
Epigraphik, p. 67, which represents the marriage of the Persian religion
(din MazdianiS) to the god Bel.

204 SEMITIC ORIGINS

with the Semitic cult, considerably modify it. At Nippur
— assuming a Semitic element in the civilization — the
cult of the mother goddess appears to have been influenced
by such a foreign element, since as early as 3800 B.C. Enlil
was not only of the masculine gender, but had also a fe-
male counterpart, which was simply his own reflection. A
transformation, which appears not to have been complete
at Eridu till some two thousand years later, seems to have
occurred at Nippur before the dawn of history. The pro-
cess was therefore probably hastened by fusion with a
foreign god.^

When our written records begin, the Semites appear to
be everywhere dominant in Babylonia ; but, as we have
seen, the phenomena of the religion compel us to postulate
a prehistoric kingdom of Nippur, which dominated much
of the surrounding country. This was probably a Sumer-
ian kingdom, into which in its later years a large Semitic
element was infused. If Enlil was originally a Sumerian
god, and Nippur the head of a Sumerian kingdom, the two
forces, religious and political, were then present which in
combination would give Enlil the place at the head of all
the pantheons which he occupied in the later religious
history.

A large Semitic element also entered into the concep-
tions of Enlil and Ninlil at Nippur. This is shown by
the following facts : 1. A great variety of phallic symbols
were found at Nippur in all the levels of the mound back
to 4000 B.C., or earlier.2 These symbols are the natural
symbols of a cult like the Ishtar cult, but do not grow so
naturally out of a purely agricultural civilization. We
cannot go astray, therefore, in regarding them as the prod-
uct of a Semitic element of thought, which entered into
the worship and life at Nippur.

1 The presence of a non-Semitic element at Nippur is confirmed by the
faces on the votive tablet of TJr-Enlil, which is of about the same age as
the inscriptions of Enshagkushanna, or perhaps a little older. See Hil-
precht, OBI., pi. xvi, and Professor Cope's note, ibid., Pt. 2, p. 48, n. 1.

2 Peters, Nippur, Vol. II, p. 236 ff.

TRANSFORMATIONS IN BABYLONIA 205

An old bilingual incantation contains the following
Sumerian expressions: AMA A-A dingir EN-LIL, and
AMA A-A dingir NIN-LIL, i.e. "the mother-father
Enlil," and "the mother-father Ninlil."i The point of
this expression is not simply that Enlil and Ninlil were
thought of as a pair of parents,^ but that the qualities of
father and mother both are actually attributed to both
Enlil and Ninlil.^ This points to the presence at Nippur
of a mother goddess who for a time almost monopolized
the thoughts of the worshippers, and who was gradually
fused with a masculine deity, with the result that for a
time, as in Lugaltarsi's inscription,* both masculine and
feminine qualities were attributed to the same deity.
Enlil was, therefore, probably originally a Sumerian earth
goddess, who by the warlike character of his worshippers
was transformed to a god, and later given a consort. The
confusion of sex probably did not last as long as in the
case of the deities Ea and Sin, which were of almost pure
Semitic origin, but this little expression in the incantation,
a fossil from past strata of thought, has transmitted to us
the evidence of its existence. By the time when written
records began Enlil and Ninlil were fairly well defined,
though even then a Semite sometimes addressed the whole
deity under the name of Ninlil.^ Another point, though
it is, as we have seen, indeterminate, is found in the fact
that Enlil and Ninlil are in an old bilingual hymn repre-
sented under the symbols of an ox and a cow, like Sin and
Ishtar.s

1 IV R., 1, col. ii, 11. 23-28.

2 Delitzsoh, Assyrisches Worterbuch, p. 20.

8 Cf. my article "An Androgynous Babylonian Divinity," in JAOS.,
Vol. XXI,2 p. 186 ff.

* See above, p. 181.

6 Cf. Winckler's Untersuchungen, p. 157, No. 9 ; KB., Vol. III,i p. 69,
and Radau, op. cit, p. 37, and p. 125, n.

s See Reisner's Snmerisch-babylonische Hymnennach Thontafeln grie-
chischer Zeit in the Berlin Museum's MUtheilungen, Heft X, p. 19,
11. 71-74 ; cf. the translation in Dr. Banks's dissertation, Sumerisch-baby-
lonische Hymnen, Leipzig, 1897, p. 23.

206 SEMITIC ORIGINS

The fusion of Semitic and non-Semitic elements simply-
hastened, as has been pointed out, the evolution which the
processes of social transformation carried on more slowly
elsewhere.

The commanding place which Enlil held in the Baby-
lonian pantheon in the earliest period is illustrated by the
way in which Enshagkushanna before 4500 B.C. presented
to him, as noted above, the spoil of his war with Kish,
although so far as appears the seat of Enshagkushanna's
government was in the south; and also, by the way
Eannadu of Shirpurla some three centuries later claims
that Enlil (not Ningursu) gave him victory over the
people of Gishban.i

The view we have been led to take of these gods throws
light on the place which Enlil and Ea afterwards held as
the two most prominent members of the triad, Anu, Bel
(Enlil), and Ea. In this triad, Bel was the god of the
earth, and Ea of the deep. Anu was in part an abstrac-
tion added at a later time to represent the third most obvi-
ous part of the universe. Bel (Enlil), the old god of the
country, though largely permeated by Semitic conceptions,
naturally took the lead, because the Sumerian kingdom
antedated the Semitic ; while Ea, the oldest Semitic god
in Babylonia, whose coming brought the artificial cul-
ture of the date-palm and infused new elements into the
civilization, whose home was on the shore of the great
water, assumed naturally a place of importance next to
Bel. When the two were united in the first triad, the lead-
ing Sumerian and the leading Semitic deities, whose hosts
had no doubt in the earlier days struggled in many a
bloody conflict, were brought into harmonious accord.

Because of a possible connection with the pantheon of

1 CTBM., Ft. VII, No. 23580, col. ii, 11. 1-7. It reads : (1) E-AN-
NA-DU MEN (2) SA UMUN GAL (3) dingir EN-LIL-LAL (4) E-NA
SUM (5) NAM-E-NA-TA KUD (6) GAL GIS H-B AN-/ri-KID (7) E-
AN-NA-DXJ-RA, i.e., "Eannadu am I. The temple of the great lord
Enlil, its greatness I established. On account of its greatness he subdued
the men of Gishban to Eannadu."

TRANSFORMATIONS IN BABYLONIA 207

Shirpurla, the next deity to be considered is the god Mar-
duk of Babylon, with his spouse Sarpanit. BalP and
Hommel ^ have suggested, in consequence of a passage in
an old hymn which identifies Gishgalla with Babylon,^
that the Gishgalla of the kingdom of Shirpurla was really
the same place as the Babylon of later history. As
Amiaud pointed* out, Gudea speaks of the whole of Shir-
purla as a city,^ a fact which precludes the possibility that
one of its quarters was as far away as Babylon. The
theory also encounters other objections which are equally
fatal to it. It is probable that the ideogram which Hom-
mel and Amiaud read Gishgalla should be read Erim.^
Gishgalla is probably another sign.'^ Eannadu, about
4100 B.C., tells us of his conquest over a city, the name
of which he represents by this latter ideogram, and which
is probably, therefore, to be read Gishgalla.^ As he calls
himself king of Shirpurla, it was clearly a town outside
of that place. He gives us no indication of where it was
situated, but from the fact that the ideogram Gishgalla
also denoted in later times the direction " south," ^ it is

1 FSB A., Vol. XV, p. 53 ff.

'^ FSBA., Vol. XV, p. 108 ff.

« IV R., 46.

* Becords of the Past, New Series, Vol. I, p. 43.

^ De Sarzec's Decouvertes, pi. 14, col. 1, 11. 14, 15.

« Cf. Thureau Dangin, Becherches, No. 359. While he there reads the
sign GISHGAL, he reads BRIM in Bevue semitique, Vol. V, p. 67. As he
himself points out (Becherches, No. 361), another sign is really equivalent
to GISHGAL, and the two signs cannot he identical. This one, as Jensen
suggested (KB., Vol. IIF, p. 3 ff.), is probably to be read ERIM (Briin-
now's List, No. 949), at least provisionally.

' Thureau Dangin's Becherches, No. 361, and Brunnow's List, No. 938.

8 Cf. the four texts in GTBM., Pt. IX, pis. 1 and 2, col. ii of each text ;
cf. also Galet A, Bev. d'assyr., Vol. IV, pi. 1, col. ill, 11. 17-19 ; Cf . Thureau
Dangin in Bev. sem.. Vol. V, p. 68, and Radau, op. cit., p. 84. The
passage reads: TU-SU-BI SUM kur ELAM Jci, TU-SU-BI SUM GIS-
GALLA ki, TU-SU-BI SUM GIS-BAN ki, TU-SU-BI SUM URU ki ;
i.e., " Into his power was given Elam ; into his power was given Gishgalla ;
into his power was given Gishban ; into his power was given Ur."

9 See Brunnow's List, No. 947. It probably acquired this meaning, as
the word Negebh did in Hebrew, by being a place southward of some

208 SEMITIC ORIGINS

altogether improbable that Gishgalla was as far north
as Babylon. It may be that Babylon was a colony of this
Gishgalla. That would afford only an indirect connection
with Shirpurla, — Gishgalla being not a part of Shirpurla
like Erim, but an independent town conquered by Shir-
purla in the historical period.

It is clear, therefore, that Hommel's identification of
Babylon with Gishgalla, even if interpreted to mean that
Babylon was a colony of the latter, will throw little light
on the nature of the god Marduk, since we have no infor-
mation whatever as to the gods of Gishgalla. If there be
any connection between the two, the deity of Babylon
might throw light on the religion of Gishgalla, but not
that of Gishgalla upon Babylon.

From other considerations, however, it can be shown that
Marduk is in all probability a Semitic god, evolved, like the
Semitic gods already discussed, out of a preceding Ishtar.
The considerations which support this view are as follows : —

1. Marduk is called in the hymns "the life giver," i
" possessor of the foundation of life,"^ and is asked to give
life.^ In other words, he is, like Ishtar, a deity of life.
2. Nebuchadnezzar tells us* that the Zagmukhu, which
in the time of Gudea was, as we have seen, a festival of
Bau, was at Babylon a festival of Marduk. It is not
necessary to suppose with Jastrow ^ that it was transferred
to Marduk ; this festival of the yeaning time in spring is
another link connecting Marduk with the Ishtar from
which he sprang.^ 3. Marduk comes first to our knowl-
edge in the inscriptions of Sumula-ilu and Khammurabi,

other important place in Babylonia, so that Gishgallaward came to mean
southward.

iIV R., 29, No. 1, Eev. 11. 5, 6 (cf. Sayce, Hihhert Lectures, p. 502).

^Ibid., Obv. 1. 38 (cf. Sayce, ibid., p. 501).

»IV R., 18, No. 2, Rev. 1. 12 (cf. Sayce, op. cit., p. 489).

*I R., 54, col. ii, 54 fi. (cf. Jensen, Kosmologie, p. 84 fl., and KB.,
Vol. Ill 2, p. 15).

^ Beligion of Babylonia and Assyria, pp. 121 , 631.

6 See above, p. 109 ff.

TRANSFORMATIONS IN BABYLONIA 209

kings who belonged, as all recognize, to a Semitic dynasty.
Their god would be a Semitic god, so that it is in conse-
quence more than probable that Marduk is developed from
the Semitic mother goddess like other Semitic deities.

From the time of Khammurabi Marduk was the chief
deity in the eyes of all Babjdonians. They worshipped
other gods as the inhabitants of other cities had done, but
unlike them, they practically placed Marduk, not Enlil,
first. Jastrow has already pointed out^ how, in conse-
quence of this, Marduk absorbed in time many of the
attributes of Bel (Enlil) and even of Ea. This movement
perhaps had its origin in a myth that Marduk was the son
of Ea.2 Sayce infers from this^ that Babylon was origi-
nally a colony of Eridu. This can hardly have been the
case, for, as will be pointed out below,* the myth was
probably in the first place a myth of Nabu which Marduk
absorbed. The name Marduk is with some probability
explained as "young" or "early sun,"^ i.e., "child of the
day,"^ and perhaps arose from the association of the
primitive Semitic goddess at Babylon with the sun.

Marduk had a consort, Sarpanitum, who first appears in
the reign of Sumula-ilu about 2360 B.C.''' Her name,
according to Delitzsch,^ comes, like that of Marduk, from
her solar character, and means "silver brightness."
In the historical inscriptions she appears to have been

1 Meligion of Babylon and Assyria, p. 117 ff.

2Cf. Winckler, Vntersuckungen, p. 140, and KB., Vol. IIP, p. 131.

' Hihhert Lectures, p. 104.

* See below, p. 212, and on the subjugation of Nabu to Marduk, Jastrow,
op. cit., of. 126 ff.

6 Sayoe, op. cit. , p. 98, Jensen, Kosmologie, p. 88, and Jastrow, op. cit.,
p. 119.

6Delitzsoli, BA., Vol. II, p. 623 n.

'King's Letters and Inscriptions of HammuraM, No. 101, ool. i, 1. 41 ;
cf. Vol. Ill, p. 218 ff.

' lUd., Contra, cf. Hal^vy, Becherches critique, p. 260, and Muss-
Amolt in JBL., Vol. XI, p. 165. In consequence of a folk etymology
the name of the goddess was sometimes written Ziru-bani-ti, or "creator
of seed." (Cf. 11 R., 67^2.)
p

210 SEMITIC ORIGINS

little more than a reflection of Marduk, but that she was
originally more than that appears from the fact that
NeWchadnezzar appeals to her as the goddess of child-
bearing.i That she always played a large role in the
popular imagination is shown by the fact that in the
middle Babylonian period the image of a nude goddess
holding her breasts was very popular in Babylonian art.^
These are in all probability representations of Sarpanit,
and indicate that in the rise of the god Marduk the femi-
nine side of the old mother goddess lost nothing of her
popularity. This is further confirmed by what Herodotus
and Strabo tell us of her service by the women of Baby-
lon.^ Sarpanitum was then the feminine counterpart of
Marduk, as Damkina was the feminine counterpart of Ea.
The pair at Babylon, as at Eridu, were probably produced
by the differentiation of the old mother goddess Ishtar.
If Babylon were a colony of Gishgalla, this conclusion
would involve the view that the latter city was also a
Semitic settlement.

Another god the origin of whose worship may with
some plausibility be traced through Shirpurla is Nabu,
the god of the city Borsippa. Hommel has suggested*
that the Kinnir, which is equated with Borsippa in the
hymn which calls Babylon Gishgalla,^ is the same town as
that mentioned in the inscription of Ur-Bau as Kinunir,^
which Hommel declares was situated in Gishgalla. The
enthusiasm of a discoverer has here led Hommel into a
slight error, for Ur-Bau does not say that Kinunir was

1 Oppert, Expedition en Mesopotamie, Vol. II, p. 295 ; Eebraica, Vol.
X, pp. 18, 19.

2 Cf. Ward in American Journal of Archaeology, 1900, pp. 291, 292.

8 See Herodotus, Bk. I, 199; Strabo, Bk. XVI, 1, 20; Apocryphal
Epistle of Jeremiah, vs. 42, 43 ; cf. Hebraica, Vol. X, pp. 20, 21, and
J£L., Vol. X, p. 79 ff.

* P8BA., Vol. XV, p. 108.

nVR., 46, cf. 11. 15, 16.

^ De Sarzec's Decouvertes, pi. 8, col. vl, 11. 9-11. Cf. Amiaud in Becords
of the Fast, New Ser., Vol. I, p. 77, and Jensen, KB., Vol. IIJi, pp.
24, 25.

TRANSFORMATIONS IN BABYLONIA 211

situated in Gisligalla, but in Girsu. It is natural to see in
Kinnir or Borsippa a colony from Girsu. The emigrants
from Kinunir, which would seem to have been a portion
or suburb of Girsu, would of course take with them the
worship of their deity. Ur-Bau tells us that the deity of
Kinunir was the goddess Dumuzizuab^ (i.e., "the living
child of the abyss," or " the Tammuz of the deep ") . He
makes it clear that she was a goddess by calling her "lady
of Kinunir." 2

Now if we take Hommel's identification to mean that
Borsippa was a colony from this portion of Girsu, then the
goddess " Tammuz of the deep " must have been the real
deity of Borsippa out of which Nabu was developed by
processes with which we are already familiar. That Nabu
had some such genesis is made probable by an old hymn
which makes him a water god and a god of fertility, such
as we have seen Ishtar to be.^ This view is further con-
firmed by a list of gods * in which Nabu is identified with
a deity of the island Dilmun, an island in the Persian Gulf
near Bahrein." It would seem, therefore, that the people
of Kinunir brought their goddess from one of the islands
of the Persian Gulf, to which they had previously mi-
grated from Arabia, and settled in or near Girsu, and that
thence a band moved onward to Borsippa. The settlement
at Girsu was made before the time of Eannadu, for he was
acquainted with their deity. ^

The proof that Nabu originated in this way is only enough
to furnish a basis for conjecture, but in the light of the an-
alogies we have traced elsewhere it seems highly probable.

1 De Sarzec, ibid.

2 Jensen's idea that Dumuzizuab must be a god (KB., Vol. 1U\ p. 25 n.)
is, if the line of reasoning in the preceding pages be at all correct,
groundless.

» IV R., 14, No. 3, 11. 10-14. Cf. Sayce, Hibbert Lectures, p. 448.

• II R., 60, 30 ; cf. also 64, 66, and Brunnow's List, 5872.

• Cf. Sayce, Hibbert Lectures, p. 114, n. 1.

• See Bevue d^assyriologie, Vol. IV, pi. 1, col. ii, 1. 9 ; cf. also Sevue
semitique, Vol. V, p. 67, and Radau, Early Babylonian History, p. 84.

212 SEMITIC ORIGINS

The name " Living child of the deep " — Dumuzizuab,
— would naturally suggest a connection of this goddess,
and hence of Nabu, with Ea. This probably took first the
form of a myth which made Nabu the son of Ea, which
would be " child of the deep " put into slightly different
terms — a myth which, as has been suggested, was proba-
bly afterward appropriated by Marduk.^ It was probably
this genesis of Nabu and his association with Ea, the god
of wisdom, which afterward made Nabu the god of learn-
ing and wisdom to the Babylonians and Assyrians.

Nabu does not appear in extant inscriptions till the time
of Khammurabi,^ and Jastrow supposes that that monarch
tried to suppress his worship in favor of that of Marduk.^
The date of his appearance and his functions as god of
fertility and wisdom all point to such an origin as has
here been supposed.* His consort Tashmit is of still
later origin, and is clearly only a feminine counterpart
of Nabu. Her name means " hearing " or " revelation,"
and is derived from Nabu's function as god of wisdom.
Her origin is therefore quite parallel to that of Damkina
from Ea.

The worship of the god Shamash, as we know it from
the inscriptions, was the native religion of two cities of
ancient Babylonia, Larsa and Agade, or Sippar. Of the
details of this worship in either city we know very little.
At Larsa, Ur-Gur, king of Ur about 2500 B.C., repaired the
temple of Shamash,^ as did also Khammurabi of Babylon
some two hundred years later.^ The latter calls the Sha-

1 Above, p. 209.

^ KB., Vol. lili, p. 123.

5 See Jastrow's Beligion of Babylonia and Assyria, p. 125 ff.

* Jensen's endeavor to make Nabu a sun god {Kosmologie, p. 239) cer-
tainly does not explain all the functions ascribed to him. The goddess
Erua, whom Sayce (Sibbert Lectures, p. Ill ff.) connects with Sarpanit and
Tashmit, and whom Jastrow (op. cit., p. 130) supposes to be the older con-
sort of Nabu, is probably the goddess Dumuzizuab under another name.

•5 Cf. I R., No. 7 ; also KB., Vol. IIIi, p. 79.

" See King's Letters and Inscriptions of Hammurabi, No. 62.

TRANSFORMATIONS IN BABYLONIA 213

masli of Larsa "lord of heaven and earth " and "shepherd " ^
— expressions also used of the other Shamash. It is hard
to say whether the worship at Larsa or at Agade is the
older. If, as Jastrow supposes,^ antiquity and fame went
hand in hand in Babylonia, he is right in the view that the
palm of antiquity should be ascribed to Agade. In this
latter city we find the worship of Shamash and Ishtar side
by side in the inscriptions of Sargon of Agade, about
3800 B.c.^ At that early time the worship of Shamash
had begun to overshadow somewhat the worship of Ishtar,
for we find Sargon making offerings and appeals to him in
which the goddess is not included.*

A very archaic tablet in the E. A. Hoffmann collection
at the General Theological Seminary in New York, which
records the gift of a field to a deity, which has not yet been
identified, speaks of Shamash as " the lady who pours forth
brightness, the mistress." There is nothing in the tablet
except a sign which is still unidentified to indicate whence
the tablet comes ; we cannot tell, therefore, whether it re-
fers to the Shamash of Agade or of Larsa. If, however, I
interpret it correctly (a matter of some doubt in the case
of writing as old as any yet discovered), it not only records
a time when Shamash was a goddess, but shows that even
then another deity, which is possibly Ishtar, was worshipped
beside her.^ Analogy makes it probable that Shamash was

1 Cf. King, ibid., 1. 2. ^ Beligion of Babylonia and Assyria, p. 70.

3 Cf. Hilprecht's OBI., No. 1, and Kadau, op. cit., p. 167 fe.

* Cf. PSBA., Vol. VII, p. 66, and KB., Vol. Ill', pp. 100, 101 ; also
Hilprecht, OBI., No. 2, and Radau, op. cit., p. 170.

' The inscription which is unpublished I interpret as follows : —

Col. I, 1. IIIMV GANA DUK-KA DINGIR ?-B:I LAG
2. SAL-LAL-TUR

Col. II, 1. IIIMVICL URTU-NI-A SIG LIK-A

2. IIIMVICL GAL PI NER-A DA-KU GUR DIMMENA
BABBAR NIN-A TAB BAR UMUN(?)

^ '• " 3005 Bur of a field, (a bed) of clay (?), to the god of 7 presented
''^ Sallaltur. "•'• 36,050 cubits in its Akkadward side, the lower (side),
from the beginning ; ^- 36,050 cubits running along the breadth of the zig-
gurat to the side of the great terrace of Shamash, the lady, who pours
forth brightness, the mistress (?)." [Continued on p. 214.]

214 SEMITIC ORIGINS

here a goddess beside whom Ishtar was worshipped.
Whether Shamash was a Sumerian goddess and Ishtar a
Semitic, we have no means of determining. Shamash may
have been an epitliet of Ishtar which hardened, as epithets
so often did afterward, into a separate deity ; or two tribes,
a Semitic and a Sumerian, may have composed the commu-
nity from which our inscription comes, and Shamash may
have been a Sumerian corn goddess, though this latter sup-
position is not probable. However this may be, Shamash
in later times was always a god. Perhaps the Semitic
settlement at Agade was very old, or the foreign influence
there was very strong. At all events, by the time of Sargon
of Agade, about 3800 B.C., Shamash was a masculine deity.
Although Ishtar appears by his side in some of the inscrip-
tions of Sargon,^ yet he could invoke Shamash without
mentioning her. Probably, therefore, Shamash was a
Semitic deity. The fact that we have found the worship
of Shamash at Shirpurla, also a Semitic community, points
in the same direction. ^

At times his worship so overshadowed that of the
goddess that Khammurabi as well as Sargon mentions
Shamash alone,^ but an old hymn, in which the goddess is
called Malkatu,* shows us the worship of the two in con-
Col. Ill, 1. IIIMVIC E BABBAR LUG AB TAB BAR

2. IIIMVICL BURU KUR IR(?) DU(?) BAD

3. LIK-A GAR-A

4. GIR(?) SAG(?)

™' '• " 36,000 cubits (along) tlie temple of Shamash, the messenger of the
father who pours forth brightness {i.e. Sin) ; ^-36,050 cubits below(?)
the mountain where the abode(?) of Ishtar (??) is, '• to the beginning ; for
making brick. •• May he strengthen, may he bless ! "

1 Cf. OBI., No. 1, and Radau, op. cit., 168, 169.

2 See, for example, Eannadu in De Sarzeo's Decouvertes, pi. 4, bis, col.
VII, 11. 7, 8.

» See KB., Vol. Ill', pp. 106-125, and Radau, op. cit., p. 169 ff.

* See Haupt's ASKT., p. 122 ff. For translations, Zimmern's Babylo-
nische Bicsspsalmen, p. 51 ff., and Hebraica, Vol. X, p. 24 ff. On the ap-
plication of the name Malkatu to the goddess, see Schrader's article in
ZA., Vol. Ill, p.353fl.

TRANSFORMATIONS IN BABYLONIA 215

junction, and in the inscriptions of Nabonidos we see them
at the close of Babylonian history reigning as a united pair.^
Between these extremes various modifications may have
taken place. One of them we can trace. In the inscrip-
tion of Nabu-apal-iddin (about 880 B.C.), Malik and Bu-
nini^ appear to be the attendants of Shamash, who rules
above and apart from them. Here Malik was no doubt
originally an epithet of Shamash, while Bunini is perhaps
another name for Ishtar.

Ishtar at Agade probably never quite lost her identity
in Shamash,^ although at times she could be ignored.
It was perhaps this goddess to whom the name Nin-
Akkad, or " Lady of Accad," is given in an old list of
deities,* though it is possible that that title is a sur-
vival from the time when " Lady " was an epithet of
Shamash.

Not far east of Babylon, where the modern mound of
Tell-Ibrahim now is, lay the ancient city of Gudua or
Kutu ^ (Kutha), of which the tutelary deity was Nergal.^
His principal temple was called Eshidlam. How old the
city was we have no means of knowing. The worship of
Nergal first comes to light in the inscriptions of Dungi,
king of Ur about 2450 B.C., who repaired his temple.^
The god was then known as Shidlam-ta-e-a,^ or " the god
who goes forth from Eshidlam," a name which appears
later in the anipu texts published by Zimmeru.^ The

1 Cf. V R., 65, col. i, 1. 35 (also KB., Vol. IIP, pp. 110, 111), and col.
ii, 1. 12 {KB. as above, pp. 112, 113) ; also V R. 61, col. i, 11. 7 and 46 ;
col. ii, 11. 5 and 40 (cf. Hehraica, Vol. X, p. 25).

2 Cf. V R., 60, KB., Vol. nil, pp. I74_i83, and Jastrow's Beligion of
Babylonia and Assyria, p. 176 ff.

3 See the evidence collected in Hebraica, Vol. X, pp. 24, 25.

4 See III R., 66, col. iii, 1. 26, rev. col. v, 11. 27, 35.
6 See Delitzsch's Wo lag das Paradies ? p. 218.

6 Cf. II R. 61, col. ii, 1. 53, and 2 Kgs. n^«.
' CTBM., Ft. IX, No. 35389, and KB., Vol. IIH, pp. 80, 81.
8 Cf. Brunnovp, List, No. 7873.

° Cf. his Beitrdge zur Kentniss der babylonischen Beligion, pp. 149,
151, 169, 165, 169, and the corresponding plates of cuneiform text.

216 SEMITIC ORIGINS

temple was also repaired at another time by Sin-gamil,
another king of Ur.^

When Nergal appears in the syncretistic pantheons of
later times, he had been assigned the twofold function of
god of the underworld and the god of death-bringing war
and pestilence.^ Jensen,^ who is followed by Jastrow,*
believes that Nergal was originally a god of the glowing
flame of the sun, and that his destructive functions are to
be attributed to that fact. As the god of destruction, they
hold that he became the god of the underworld.

It may well be doubted whether this view will satisfac-
torily explain all the facts. The solar explanation of
deities which are ancient are, I believe, never able to lead
us to the most primitive character of the god. Men
thought of objects on the earth, and identified their gods
with them before they thought of identifying them with
anything in the far-off sky. If Nergal was the deity of
Kutha in that early time when each city was independent
and had its own god, it is certainly unlikely that they
then identified him with the glowing heat of the sun. It
is much more probable that he was then an agricultural
god, a deity of the soil and the giver of fertility. When
the gods of the various cities were grouped in a pantheon,
Enlil of Nippur, who was also a god of the soil, took pre-
cedence of Nergal, no doubt because Nippur was a more
powerful city. Nergal could of course not be assigned
the same functions, but was still connected with the earth,
though limited in his sphere by being assigned to the
underworld. As lord of the region of the dead he would
naturally be conceived as eager to people his realm, and
so become in time the god of war and pestilence — forces
which cause death. This might naturally lead also to his
identification with the glowing heat of the sun. We thus,

1 KB., Vol. im, pp. 84, 85.

2 IV R. 26, No. 1.

' Kosmologie, pp. 476-487.

* Religion of Babylonia and Assyria, p. 66 fE.

TRANSFOEMATIONS IN BABYLONIA 217

I believe, have a genesis for Nergal more probable than
that suggested by Jensen.

So far as appears Nergal was a Sumerian god. There
is no trace, in the scanty information concerning him
which has come down to us, of the peculiar characteristics
of fertility which attach to all the chief Semitic deities.
Delitzsch long ago called Kutha one of the oldest centres
of Sumerian civilization,^ and that still seems the more
probable view. This old Sumerian agricultural god was
adopted by the Semites and assigned a place in their pan-
theon as the god of the underworld. ^ The etymology of
his name is uncertain. ^

Another people whose home lay to the eastward of
Babylon across the Tigris were the Guti, sometimes
called the Suti.* A Semitic king of this country has left
us an inscription which dates from 3800 B.C. or earlier.^
In this inscription the monarch invokes the deities, Guti,
Ishtar, and Sin. From what we have already learned of
the god Sin and the religious syncretism of this period,
it is clear that this deity was not native to the Guti. Of
the other two, Ishtar is of course our old Semitic goddess.
From a list of Babylonian deities which comes to us from
the library of Assurbanipal,^ we learn that the worship of
Ishtar was maintained here in much of its primitive purity
down to a much later time. In our extant inscriptions
the god Guti does not, so far as I have observed, appear
again. This one glimpse of him makes upon one the

1 Faradies, p. 217.

2 By the transportation of Kutheans to Palestine by Sargon (2 Kgs.
17^*"'*), the worship of Nergal was introduced among the western Semites.
It seems to have spread from Samaria to Sidon, and thence was carried
by Sidonian emigrants to Athens. See CIS., Pt. I, Vol. I, No. 119.

3 See Jensen and Jastrow as cited above.

4 Cf. Delitzsch, Faradies, pp. 233-2.37.

5 Cf. Winokler, ZA., Vol. IV, p. 406 ; Hilprecht, OBI., p. 12 ff. ; Radaii,
Early Babylonian History, p. 175 ff., and Rogers, History of Babylonia
and Assyria, Vol. I, p. 359 ff.

6 III R. 66, reverse col. vi, 11. 18-26. Cf. translation and comments in
Hebraica, Vol. X, p. 26 ff.

218 SEMITIC ORIGINS

impression that he was probably the pre-Semitic god of
the country Guti, who had been adopted by the Semitic
immigrants in accord with conceptions with which we are
already familiar, and associated with their goddess.

Another of the petty states of ancient Babylonia of the
god of which we get a glimpse in the inscriptions is Gish-
ban, a place which as Thureau Dangin has shown lay just
north of the Shatt-el-Khai.^ We learn from the inscription
of Lugalzaggisi that the chief god of this place was repre-
sented in writing by the two signs SI-ELTEG(?), which
mean " the one who pours forth grain," ^ but which a much
later tablet defines as Mdaba.^ The emblem of Nidaba
was the waving grain ; for in the Gilgamish epic, the un-
kempt hair of the wild man, Eabani, is said to have grown
as luxuriously as Nidaba.* We know, therefore, that this
deity was an agricultural deity, and a giver of fertility.
Lugalzaggisi, whom Hilprecht believes to be a Semite be-
cause of Semitisms in his inscription,^ calls himself a son
brought up by this deity. There is no direct evidence in
the inscription as to whether Nidaba was masculine or
feminine, but grain deities are so often feminine, that
whether Gishban was a Semitic settlement or not, it is
probable that Nidaba was a goddess or developed out of a
goddess. The culture either of Arabia or of Mesopotamia
might, so far as we can tell, have produced this deity.
We must leave the origin of this goddess, therefore, to be
determined when further inscriptions have arisen from the
dust to throw light on her character.

The origin of the god Anu is shrouded in great ob-
scurity. From the time of Gudea^ onward, and probably

1 See Comptes rendus de V academic des inscriptions et belles-lettres,
Vol. XXIV, (1896), p. 593 ff. and Bevue d'assyriologie, Vol. IV, p. 41.
^ See Briinnow's List, Nos. 7433 and 4447.

• Brunnow, op. cit., No. 7453.

* Haupt, Nimrodepos, p. 8, 1. 37 ; cf. Jensen in KB., Vol. VI, p. 121,
who renders Nidaba by Weizen " Wheat."

6 OBI., Pt. II, p. 55.

' See Statute B (De Sarzec, Decouvertes, pis. 16-20), col. viil, 1. 45 ff.,
and KB., Vol. Ill', p. 4fi, 47.

TRANSFORMATIONS IN BABYLONIA 219

from the time of Anu-baniui,^ some eight hundred years
before Gudea, Anu in theory stood at the head of the
Babylonian pantheon. We are, however, unable to con-
nect his name with any city the political importance of
which would help to give him this commanding position.
We learn from an inscription of Nebuchadnezzar I ^ (about
1130 B.C.) that the city Der, situated on the Tigris,^ was
a city of Anu ; but this city plays no part, so far as we
know, in early Babylonian history, and the god can hardly
have been placed at the head of the pantheon in conse-
quence of its importance. It is probable, as Jastrow has
suggested,* that he was given this position as the result
of those abstract and more scholastic conceptions which
resulted in the formation of the first triad, Anu, Bel, and
Ea, the gods of heaven, earth, and the deep, of which
the god of heaven, Anu, naturally took the first place.
Jastrow supports this view by the supposition that the
heavens were not really personified as a god till about
the time of Khammurabi. He reaches this conclusion in
part because of the fact that in passages which are often
interpreted as referring to Anu the determinative for god
is not prefixed to the name of the deit3^

This latter fact does not necessarily support the view
in question, but is open to another explanation. The
name Anu was written by the sign an with a phonetic
complement. An had also as a determinative the value
dingir (ilu), and was placed before the names of gods.
When repeated it stood for the plural "gods." To write
it twice for the name of Anu would suggest to the reader
a plural, and tend to create confusion ; it may have been
omitted from the name of Anu for this reason. It is true

1 See JRecueil de traveaux, Vol. XIV, pp. 100-106, and Radau, Early
Babylonian History, p. 177 ff.

2 See Hilpreclit's Freibrief Nebuchadnezzars I, 1. 14, and Peiser in
KB, Vol. IIF, pp. 164, 165.

3 Hilprecht, OBI., No. 83, 1. 2, Assyriaca, pp. 10, 11, and Peiser, KB.,
Vol. IV, pp. 64, 65.

* Beliyion of Babylonia and Assyria, p. 80 ff.

220 SEMITIC ORIGINS

that in a number of passages in Lugalzaggisi, Gudea, etc.,^
it is possible to translate the sign as an adjective as Jas-
trow would, but in the inscription of the king of Lulubi,
who, before 3800 B.C., erected an inscribed stele in the
mountains near the modern town of Zohab,^ such is not
the case. Anu and Anat were then already deities at the
head of Anu-banini's pantheon.^ Not only so, but the
king bears the name Anu-banini ("Anu is our begetter"),
a name which suggests that some chthonic god of fer-
tility — a god originally connected with some tribe or place
— had been identified with the heavenly expanse, so that
an earthly history really lay back of this celestial deity.

The name of the Semitic king, together with the fact
that Anu and Anat stand at the head of his pantheon,
suggests the view that this pair may have been developed
out of an Ishtar at Lulubi, as Ea and Damkina were at
Eridu. This hypothesis cannot, in consequence of the
scantiness of our present information, be either proved or
disproved. It is also possible that Anu may have been
some pre-Semitic god of Lulubi, whose worship the Semites
had adopted on coming to the country ; but if so, they
had probably merged the cult of their own goddess with
him till she became Anat, so that by the time of Lulubi
the history of Anu and Anat was parallel to that of Enlil
and Ninlil at Nippur. It is at all events probable that
Anu resulted from the identification of an earthly deity
with the sky, and was at the first no more of an abstrac-
tion than Sin and Shamash were.

1 See Hilprecht, OBI., No. 87, col. i, 1. 5 ; De Sarzec, Decouvertes, pi.
13, Nos. 1, 2, col. i, 1. 3 ; col. ii, 1. 15; and pi. 13, No. 4, col. i, 1. 3. See
also Thureau Dangin in Eev. semitique, Vol. V, p. 269 ; Amiaud in Bee.
of Past, New Series, Vol. II, pp. 92, 93, and 103 ; also Eadau, op. cit., pp.
152, 202, 204, 209, 257, 267, 280, and 281. The determinative is not infre-
quently omitted, however, before the names of deities, especially in the
older inscriptions.

2 Rogers, History of Babylonia and Assyria, Vol. I, p. 360 ff. , and
Radau, op. cit., p. 177 ff.

' It is impossible in this inscription to translate in any other way thaa
as the name of a god.

TRANSFORMATIONS IN BABYLONIA 221

It seems clear from the preceding discussion that the
application of our economic-religious test to the gods of
Babylonia sheds a little light on what was Semitic in
ancient Babylonia, and what may with plausibility be
claimed as non-Semitic. The test cannot at present be
applied throughout in consequence of the fragmentary
character of the material, nor is it a test which will in all
cases yield perfectly definite results. It is one, notwith-
standing, which should be applied conjointly with linguis-
tic tests, and in the mixed problem of Babylonian origins
it proves its worth. If it leads us at times to determine
the boundaries of nationality somewhat differently than
we should from linguistic evidence alone, that is only a
tribute to its value.

About the middle of the nineteenth century B.C. the
written records of the Assyrian people whose home lay to
the north of Babylon begin. This kingdom was primarily
the dominion of the city of Ashur. The Assyrian empire,
like the Roman, resulted from the dominion of a single
city. This city was the city of the god Ashur, who thus
became the national god of the Assyrians.^ The Assyrian
was even more than most of the empires of antiquity a
well-organized fighting machine, and, as all the statements
about Ashur occur in inscriptions written after the era of
conquest began, they necessarily represent Ashur as a god
of war.2 As a local deity he must originally have pos-
sessed all the functions of a local god, among which would
be in an agricultural community those of fertility. Some
recollection of this has survived in the language of Assur-
banipal, who calls himself the offspring of Ashur and

» The name of the city Ashur appears originally to have been derived
from the name of the god. This is not so strange as Jastrow (pp. cit.,
p. 196) thinks. The same was true of NinS (see above, p. 155 ff.), and
probably Nineveh (see belovf). Eridu, Ur, and Nippur are represented
by the same ideograms as Ea, Sin, and Enlil, showing that at some time
the names of these gods and their cities were the same.

2 For a statement of this phase of Ashur, see Jastrow's Religion of
Babylonia and Assyria, p. 193 ff.

222 SEMITIC ORIGINS

Ishtar.^ With him the expression was perhaps somewhat
figurative, but it points to a primitive conception of these
gods similar to that underlying Talab Riyam and Shams
in south Arabia, whose worshippers regarded them as
their parents.^

Ashur was, so far as appears, a purely Semitic town,
and although there existed in it a temple of Anu and
Ramman^ which was built before our written records be-
gin, the worship of these gods must have been a later im-
portation than the worship of that deity for whom the
city was named. The god Ashur cannot be connected
with either of these since he is never connected with any
of the elemental powers of nature. Nineveh and Arbela
were founded by Semites who brought with them the
worship of some form of the goddess Ishtar,* and while
Ashur is probably older than either of them, it is probable
that the immigrants who founded it did the same. Haupt
suggested some years ago that the name Ishtar was derived
from the name Ashur. ^ This view we have found it im-
possible to accept, but it is possible that the reverse may
be true, and the name Ashur be derived from Ishtar. In
Assyrian the 'Ayin and 'Aleph were both so weakened as
to be at times indistinguishable, so that it only remained
to assimilate the t to the preceding sh to transform the
name of the goddess into that of the god.®

1 V R., 1, 1. Cf. KB., Vol. II., pp. 152, 153.

2 See above, p. 130.

SIR., 15, 60 ff. ; cf. KB., Vol. I, pp. 42, 43.

* See below. Chapter VI. ' See above, p. 103, n. 2.

* Such assimilation of a J to a preceding S is not infrequent in the com-
mon speech of the Babylonians and Assyrians (of. Delitzsoh, Assyr.
Gram. § 51, 2). The s was in such cases usually changed to s, but in the
name Ishtar other phonetic laws of the Babylonians and Assyrians suffer
variation, e.g., S before a dental is usually changed to I, but in the name
Ishtar the S always held its place. The SS in ASSur, if real, may be a
similar exception. It is not certain that it is real, however. If the view
of Tiele and Muss-Arnolt (p. 223, n. 1), represents as assumed below a folk
interpretation of a later time, the writing of the name may have been
changed from Assur to ASsur in the state inscriptions in accordance with

TRANSFORMATIONS IN BABYLONIA 223

Tiele ^ and Muss-Arnolt would connect the name with
the root a-sh-r which occurs both in Hebrew and Assyrian
in the sense of "be gracious, bless, cause to prosper."
Although this view is supported by the fact that the name
of the god is written by an ideogram which means good,
it is probable as Jastrow ^ suggests that he was called the
"good" as a mere epithet. It is most probable that the
epithet was applied to Ashur by a folk etymology, which
made a play upon the name as is so often done with Old
Testament names and as was done at Babylon in the case of
Sarpanit. If so, this view is really an argument in favor
of another origin of Ashur. Possible as I consider these
etymologies to be, Hommel has suggested one ^ which must
be regarded as far more probable. He takes the name
like the Assyrio-Babylonian word for sanctuary (ashirtu),
to be derived from the old 'asheras or posts which marked
the boundaries of Semitic sanctuaries. In several parts
of the Semitic world the name of the post was transferred
to the goddess,* and in one other case the goddess was in
all probability transformed into a god. Such really seems
to have been the course of events in Assyria. This view
is supported by the fact that Khammurabi seems to have
known such a goddess,* and it affords a simple and satis-
factory etymology for Ashur.

The general development of purely Semitic deities from
the primitive mother goddess as a starting-point estab-
lishes a strong probability that Ashur was a transformed
Ishtar. In favor of this view is the fact that there is but

this interpretation. That the speech of the people was not always in simi-
lar cases represented in the writing Delitzsch admits (Gram. p. 119). I
regard another origin, however, as far more probable. For another folk
interpretation of the name see below, p. 224, n. 3.

1 Babylonisch-Assyrische Geschichte, p. 533. So Muss-Arnolt 3and-
worterhuch, p. 118.

2 Religion of Babylonia and Assyria, p. 196 ff.
8 Aufsatze und Abbandlungen, Vol. II, p. 209.

* See below Chapter VI.

* Cf. King's Letters and Inscriptions of Hammurabi, No. 66.

224 SEMITIC ORIGINS

one trace of an Ishtar of the city of Ashur, and that is a
very late one, capable of another interpretation. Assur-
banipal speaks of " Ishtar the Assyrian," ^ but in reality
he probably refers to the goddess of Nineveh, who had by
his time long been associated with the god Ashur. Nine-
veh was a part of the Assyrian dominion long before the
earliest of our extant inscriptions, and it is probable that
the worship of its goddess had been united with that of
Ashur, so that Raman-nirari I and Tiglath-pileser I, when
they refer to Ishtar, mean the goddess of Nineveh. ^

Probably, therefore, the original goddess of the city of
Ashur was transformed into a god before the dawn of the
historical period, and after Nineveh had been conquered
its goddess became, through the operation of those laws
of syncretism with which we are already so familiar, the
spouse of the god Ashur. ^ The deity whose worship next
to that of Ishtar was most widely extended over the Se-
mitic world was the god known in Assyria as Ramman and
among the Aramaeans as Hadad (in cuneiform Addu), or
Rimmon. The most widely recognized function of this
god caused him to be regarded as the god of thunder,
lightning, wind, and storm, though as we shall see other
attributes were not lacking in the minds of some of his
worshippers.

1 V R., 1, 65 ff.; cf. KB., Vol. II, pp. 158, 159, and Hebraica, Vol. IX,
pp. 156, 157.

2 This point was not clear to me when the article on the " Ishtar Cult "
was written. The classification of the Assyrian material adopted in that
article (Hebraica, Vol. IX, p. 131) was I now think a mistake. Its re-
sults were not very far-reaching, as it only led to the assignment of four
or five allusions to the city of Assur which belonged to Nineveh.

2 It seems probable as Jensen and others have suggested (see Jensen,
Kosmologie, p. 275, and ZA., Vol. I, p. 1 ff. ; Delitzsoh, Wellsclidpfrmg-
sepos, p. 94 ; and Jastrow, op. cit. , p. 197), that tlie god Ansliar who plays
a prominent part in our present version of the Babylonian creation epic is
intended for the god Ashur, and is introduced as a compliment to Assyria,
An-shar being a dissimilation of ASsur. If this be true, however, it proba-
bly does not help us with the real etymology of the name Ashur, but is a
folk etymology similar to the oue discussed above, p. 223 and 222, n. 4.

TRANSFORMATIONS IN BABYLONIA 225

The name by which this deity was known in the older
Babylonian period is in dispute.^ His name is written by
the ideogram IM, and Thureau Dangin may be right in
holding that in the oldest period it was pronounced " Im-
meru."^ However this may be, he was certainly called by
the Assyrians Ramman. Material recently made accessi-
ble to scholars makes it clear that the worship of this god
is of great antiquity in Babylonia. We do not know in
what locality his worship first originated, but he was in-
voked by Anu-banini before 3800 B.c.,^ and is coupled by
that monarch with the goddess Ishtar. The same god
also appears as a deity of popular worship on tablets of
the time of Bur-Sin, king of Ur.* He must therefore
have had a long career in Babylonia before the time of
Khammurabi, although the material so far recovered does
not enable us to trace it. All that we know of his nature
in this early time is that the ideogram IM, " wind," indi-
cates that he was connected in some way with the weather.
By Khammurabi he was worshipped, and was associated
with Shamash.^ Shamsu-iluna built a fortress or town to
him.^ Later in the Kassite period he became very popu-
lar, and several of the kings bore names which ascribed
honor to him. At this time a second triad of gods appears,
composed of Sin, Shamash, and Ramman.'

1 That he was called by several names appears from the tablet published
by Bezold, PSBA., Vol. XI, pp. 173, 174, and pi. 1. For discussions as to
the name, see Hilprecht, Assyriaca, p. 76 ff.; Oppert, Comptes rendtis de
Vacademie des inscriptions et belles-lettres, June, 1893 ; Jour, asiatique,
1895, pp. 39.3-396, and ZA., Vol. IX, pp. 310-314 ; Thureau Dangin, Jour,
asiatique, 1895, pp. 385-393; Jastrow, AJSL., Vol. XII, pp. 159-162;
and Beligion of Babylonia and Assyria, p. 156 ff.

2 He was also called Mer in Babylonia and Bir in Syria. Cf. Hilprecht,
Assyriaca, p. 77 ft.

3 See Becueil de traveaux. Vol. XIV, pp. 100-106, and Radau, Early
Bahylonian History, p. 177.

« Cf. Radau, op. cit., pp. 327, 353, 1. 33 ; 427, 1. 6, and 429.
6 Cf. KB., Vol. IIP, pp. 112, 113.
« KB., Vol. nil, pp. 132, 133.

' See any list of the kings of the third Babylonian dynasty, and Belser
in BA,, Vol. II, p. 201, col. vi, 1. 8.
Q

226 SEMITIC ORIGINS

In Assyria the worship of Ramman goes back to pre-
historic times,^ and he was very popular in the historic
period as the god of storm, lightning, and thunder, who
helped to overthrow the enemies of his worshippers.^ Tig-
lath-pileser I once refers to him as the god of the " west
country,"^ which shows that he identified him with the
Aramaean deity. It is hardly probable that Ramman
was born on Assyrian soil. It is more probable that his
worship was carried thither by Babylonian or Aramaean
immigrants, preferably the former.

In the El-Amarna letters from Syria and Palestine the
ideogram IM is used to represent the name of a Syrian
god, which is at times spelled syllabically as Ad-di,*^ and
once as Ha-da-di.^ The same writing occurs centuries
later in a contract written in Babylonia for an Aramaean
immigrant.^ These passages equate the Syrian god Hadad
with the Babylonian-Assyrian god Ramman. The equa-
tion was a most natural one to make, as the names Ramman
and Hadad both appear to have meant "Thunderer."'

The worship of this god in Damascus is known to us
through the Old Testament, where his name usually ap-
pears as Hadad,^ but once a corruption of the Assyrian
form occurs as Rimmon.* It appears from the obelisk

1 I E., 15, 71 ff. Cf. KB., Vol. I, pp. 42, 43.

2 Cf. Tiglath-pileser I in I R., 9, 9 ff. and 78 ff.; also KB., Vol. I, pp.
16-19. Assurnasirpal exhibits the popularity of Ramman by calling him
" The mightiest of the gods " {KB. , Vol. I, pp. 116-117) .

8 I R., 14, 87 ; KB., Vol. I, pp. 38, 39.

' Cf. KB., Vol. V, Nos. 333, io4i, and 85'. The name is written by
Assurbanipal Da-ad-da (VR., 9, 2; cf. ^5., Vol. II, pp. 222, 223).

^ KB., Yoi.Y, No. 881.

« See TSBA., Vol. VIII, p. 282 ff., and Sp. 41; also Strassmaier's
Nabonidos, No. 356.

' For " Ramman " cf. Delitzsch, HWB., p. 624, and Jastrow in AJSL.,
p. 160 ff.; for "Hadad " cf. Buhl in Gesenius's Handworterbuch, 13th ed.,
p. 191 ; the Brown-Robinson-Gesenius Lexicon, p. 212, and Hoffmann,
ZA., Vol. XI, p. 227.

8 See 1 Kgs. 15'8.2o, 20p«««™, and 2 Kgs., e^^.^to,^ where it appears as
the divine element in the name of the Icing of Damascus.

° 2 Kgs. 5'6. Cf. Baudissin's discussion, Studien zur semitisehen
Belu/ionsgescliichte, Vol. I, .308 ff.

TRANSFORMATIONS IN BABYLONIA 227

inscription of Shalmeneser that he was also known as
" Bir." ^ He seems to have been the chief deity of Dar
masons, as well as of the Aramseans generally.^ Two
passages in the El-Amarna letters prove that he was re-
garded in Assyria as the majestic thunderer who over-
whelmed enemies.* His worship would seem to have been
carried in this period to several places in Palestine by
Aramaean immigrants, where in after centuries places
bearing the name of " Rimmon " attest the fact that the
worship of a deity bearing that name had once held sway.*

The excavations at Zendchirli, in the extreme north of
Syria, have brought to light the statue of the god Hadad,
together with an inscription partly in his praise written
by a king of the eighth century B.C. His worship would
seem to have extended wherever Aramaeans went. In
their migrations they carried it to Babylonia, as already
noted, and the name of the god has been found in an Ara-
maic inscription as far south as Telloh.^ In Egypt it has
been found in the vicinity of Memphis,^ while the worship
penetrated north Arabia at Hegra,^ and into south Ara-
bia,^ where the god was known as Rimmon ; while as the
cult of Hadad it is also found in both the northern and
southern parts of that peninsula.^

To determine the origin of such a god from the general
point of view of our preceding discussion would seem at
first glance a difficult matter. The disciples of Max Miil-
ler, who take every deity for the personification of some

1 See Abel and Winckler's Keilschrifttexte, p. 8, 1. 59, and p. 9, 1. 88 ;
also Hilprecht, Assyriaca, p. 77.

2 See the name of the king of Zobah in 2 Sam. 8^^^^
8 KB., Vol. V, Nos. 149" and 150'.

* Josh. 1532, jud. 20", 2113, and Zech. 12".

5 CIS., Pt. II, Vol. I, No. 72.

6 CIS., Pt. II, Vol. I, No. 124.
' CIS., Pt. II, Vol. I, No. 117.

8 Cf. Glaser's Die Abessinier in Arabien und Afrika, p. 105.

' See HaMvy, Melanges de critique, p. 424 ; Bevue semitique. Vol. 11,
p. 21 ; Winokler, JJntersuchungen, p. 69, n.; Noldeke, ZDMQ., Vol. XLI,
p. 712 ; and Wellhausen, Beste arab. Heidentums, 2d ed., p. 55.

228 SEMITIC OlilGINS

natural element, might be thought to have here a clear
case of a pure and simple storm god.

There are not lacking, nevertheless, conceptions con-
nected with this god which such a theory is powerless to
explain. The name of an Old Testament city, Ain-Rim-
mon, or "Fountain of Rimmon,"i proves that the god was
once connected with a spring, while the request which Naa-
man made of Elisha concerning Yah we ^ suggests that he
was accustomed to connect his own god, Rimmon, with
the soil. Panamu of Zendchirli also calls Hadad "Baal
of water,"^ which shows that he regarded him as a Semitic
Baal. The proper name, Ben-Hadad-nathan, or " Hadad
has created a son,"* of which the Biblical name Ben- Hadad
is an abbreviation,^ proves him to have been connected
with animal fertility as well as vegetable productiveness.
It would thus seem that Hadad had been an earthly Baal
before he became the god of storms and thunder. Jastrow
has pointed out^ that in Assyria Ramman was at times
identified with Shamash, who was a god of fertility, so
that it is probable that a similar earthly history lay back
of him there.^ This is rendered practically certain by
an old Babylonian hymn, which caUs Ramman "lord of
wells." 8

1 Josh. 15^2. Cf. recent commentaries.

2 2 Kgs. 5".

^ Konigliche Museen ztc Berlin, — Mittheilungen aus dem orientalischen
Sammlungen, Heft XI, Taf. vii, 1. 1, or Lidzbarski's Handbuch der nord-
semitische Epigraphik, Taf. xxii, 1. 1. Cf. Cook's Glossary of Aramaic
Inscriptions, p. 32.

* See references, p. 226, n. 6.

5 This fact was not clear to me in 1895 ; cf. JBL., Vol. XV, p. 175.

» Beligion of Babylonia and Assyria, p. 211.

' This view receives some corroboration from the fact that in one of the
old hymns publishfid by Heisner (Berlin, Mittheilxmgen, Heft X, p. 23,
1. 10 ; cf, Banks's Snm.-Bab. Hymnen, p. 25), Eamman is described as a
wild ox. The bull as already noted is a symbol of the gods of fertility in
agricultural communities, and if Eamman had once been such a god, it
would be very natural when he had become the destructive god of storms
to change the domestic ox which symbolized him into a wild ox.

* Translated by Sayoe, Hibbert Lectures, p. 530 ff.

TRANSFORMATIONS IN BABYLONIA 229

Indeed, the steps by which a god of the soil — a giver of
grain — became a storm god are very clear. As the god
of fertility he was naturally the giver of rain, and as the
god of rain he would become the god of thunder and
lightning. When, as in Babylonia and Assyria political
union caused the formation of the gods into a pantheon,
the more active functions of production were thought to
be presided over by other gods, the rain god would in time
be associated with the more violent manifestations of his
power, and become the god of storm and destruction.

If then a god of fertility — a Baal — were a stage in
the development of Ramman-Hadad, it is highly probable
in view of the many cases of transformation which have
been already traced that this Baal was in turn a metamor-
phosis of the primitive goddess of fertility. Thus the
evolution of the storm god probably followed the same
course as that taken by other Semitic deities.

This evolution probably went on independently in
Babylonia and am^ong the Aramaeans.^ The god wor-
shipped by Anu-banini 3800 B.C. can hardly have been
affected by Aramaic influence, and we have no reason to
suppose that Babylonian influence seriously affected the
development of Hadad. In Assyria two waves of his
worship meet and unite, — one from Babylonia and one
from the West.

The worship of the god Dagan like that of Ramman is
found in both the East and the West, — in Babylonia and
Assyria on the one hand and in Palestine on the other.
It is found about 2500 B.C. in the name of a prince of
Nippur,^ and on a tablet from the time of the sovereignty
of Ur which is perhaps earlier still.^ It also appears in
an inscription of Khammurabi.* In Assyria the cult

1 Of course In Babylonia some Sumerian influences may have hastened
the process of evolution.

s I R., 2, No. 5; KB., Vol. Ill, pp. 86, 87.
" Radau, Early Babylonian History, p. 261.
1 Schrader, KAT.'^, p. 181.

230 SEMITIC OKIGINS

seems to have been established before the written records
of that country begin, for we find it as an element in a
proper name at the very dawn of history. ^ Dagan seems
to have been worshipped in Assyria down to the eighth
century B.C. as the inscriptions of Assurnasirpal, Shamshi-
Ramman, and Sargon show.^ According to the Hebrew
version of Tobit it continued till the time of Sennacherib,^
but this cannot be regarded as reliable evidence.

The El-Amarna letters attest the presence of Dagan in
Palestine in the fifteenth century B.C.,* while we learn
from the Old Testament that he was the god of Gaza^
and Ashdod,® and the inscriptions of Sennacherib make it
probable that he was the god of the Philistines generally.'^
He appears also to have been once worshipped near Nablus*
and in the neighborhood of Jericho.*

Various theories of the origin and nature of Dagan
have been propounded. Rashi advanced the idea that his
name was derived from the Hebrew dag, fish, and that
Dagon was a fish god.^" In recent times attempts have
been made to strengthen this view by comparing the
Babylonian pictures of the fish god Ea, but the compari-

1 KB., Vol. I, pp. 42, 43 ; ZA., Vol. V, p. 79, and Hebraica, Vol. IX,
p. 132.

2 See references in Jensen's Kosmologie, p. 452.
' See Neubauer's edition, p. 20.

* Cf. the name Dagan-takala in KB., Vol. V, Nos. 215, 216.
6 Jud. 1623.

«2 Sam. 5^*; 1 Mace, lOSS'*, 11*, and Josephus, Antiquities,
xiii, 4^.

' KB., Vol. II, pp. 92, 93; cf. Schrader, KAT.'^, p. 181 ; and Delitzsch,
Faradies, p. 289.

* See Bait Dejan on Pal. Expl. Funds, map seven miles east of Nablus,
and G. A Smith's Historical Geography, p. 332, n.

' Josephus, Antiquities, xiii, 8\ and Jewish Wars, i, 2'.
1" See Moore's article "Dagon" in Encyc. Bib., and Jules RouTier
in Jour, asiat., September, October, 1900, p. 347 ft. 'iiSdKuv, who was
according to Eusebius a Babylonian fish god, is also compared. Cf.
Schrader, KAT.'', p. 182 ; and Jensen, Kosmologie, p. 450. Eusebius is,
however, too late to count for much when unsupported. His Odakon, per-
haps, is for a corruption of Oannes.

TRANSFORMATIONS IN BABYLONIA 231

son is inapt, since the Dagan of the Babylonians was
quite a different god from Ea.

Philo Biblos took the name from the Hebrew ddgdn
"grain" and regarded Dagon as an agricultural deity. ^
This view, though rejected by many modern scholars, prob-
ably comes nearer to the truth than the former one. The
real nature of the god cannot be determined, however,
without taking into account the evidence from both the
East and the West.

In Babylonia Dagan was associated with Bel, the god of
the earth, and his cult would seem ultimately to have been
merged into that of Bel.^ Dagan must therefore have
been a god of the earth like Bel, or in other words he was
a Baal ^ — a god of the soil. Jensen holds that he was a
Semitic deity,* and believes with Jastrow ^ that the Baby-
lonian god is closely related to the god of Philistia.

The course of evolution by which the great Semitic
deities were produced leads us to suspect that the Semitic
Baal called Dagon was, like the others of his kind, devel-
oped out of a still earlier mother goddess in some sheltered
nook at a time when intercommunication had not pro-
duced religious syncretism. It is a difficult matter to
determine where the sheltered nook which formed the
earliest habitat of Dagon was situated. His worship may
have originated in Babylonia, whence it was carried to
Assyria in prehistoric days and to Palestine before the
El-Amarna period. It can hardly have been the native
religion of the Philistines before their coming to Palestine,
but must have been adopted by them because Dagon was
the god of their newly acquired home.^

1 See Sanchoniathontis Fragmenta, ed. Orelli.

' III R., 68, 21 0, d, and Jensen, Kosmologie, p. 453.

» So Jensen, op. cit., p. 456. * IMd., p. 455.

^ Religion of Babylonia and Assyria, p. 208.

6 Whence the Philistines came we do not know ; perhaps from Asia
Minor. Cf. the article of W. Max Miiller, " Die Urheimat der Philister,"
in Mitteilungen der vorderasiatischen Gesellschaft, Vol. V, Heft 2, pp.
1-13.

232 SEMITIC ORIGINS

Another view is that Dagon was an Aramsean god
whose worship radiated from the highlands between Pales-
tine and Mesopotamia to the countries on both sides. ^ In
view of the early appearance of the name in Babylonia
this theory encounters grave difficulties. We cannot at
present pronounce definitely upon the matter, but must
patiently wait for the appearance of further material. It
seems probable, however, that Dagon was a Baal developed
at some point on Babylonian soil out of the primitive
Semitic cult, and that thence his worship was diffused by
emigration to Assyria and Palestine. ^

^ So Jastrow, op. cit., p. 208, and Sayce, Hibbert Lectures, p. 188.
Sayce infers from Sajrgon's declaration that tie had extended his protec-
tion over Harran and according to the ordinance of Anu and Dagon
written down their laws, that Dagon was especially connected with
Harran — a conclusion which seems no more necessary for Dagon than for
Anu.

2 The god Nusku was a fire god (cf . IV R. , 26, No. 3, Sayce, Hibbert
Lectures, p. 497). His origin is obscure. He was worshipped in the
Assyrian period especially by Shalmeueser II and Assurbanipal.

Ninib (Adar ?) was the same as or a development from Ningirsu, (cf.
II R., 54, 74, and Briinnow, List, No. 10994), so that his origin has
already been discussed.

In the oasis of Palmyra, some 150 miles northeast of Damascus, no
special male deity seems, so far as the inscriptions indicate, to have been
developed, but the inscriptions show that the worship of Babylonian and
Syrian gods was brought here by immigrants from different directions.
The Babylonian Bel (De Vogiie, Syrie centrale [Palmyre] Nos. 117,
140), Shamash (No. 8), and the Syrian Baal (Nos. 16, 73) all appear.

CHAPTER VI

SURVIVALS

Having briefly traced in the two preceding chapters
the transformations of the primitive Semitic goddess in the
different parts of the Semitic world, something should now
be said of the survivals of her cult. These have incident-
ally been already introduced in part at various points of
the argument as they were needed, but have not all of
them been adequately treated. They merit a brief, con-
nected discussion. We shall begin with Arabia, the primi-
tive Semitic home.

A clear case of survival here, though under a different
name from that of the primitive divine mother, is the god-
dess Al-Lat, whose worship can be traced in several parts
of Arabia. At Taif, to the south of Mecca, it flourished
among the Thaqif,i where an old stone nosh or masseba
of her still remains, and was seen by Doughty. ^ At Sal-
khad her cult can be distinctly traced in the Nabathsean
inscriptions.^ We learn that a temple was built to her
there at one time, and at others a candlestick and a nosh
were consecrated to her. In these inscriptions she is called
the "mother of the gods." The other gods of the place
were Dhu-'l-Shara and Manutu.* The god of whom she was
especially the mother appears from a passage in Epipha-
nius,^ who vouches for the presence of her worship at Petra

1 Cf. Ibn Kutaiba, p. 60, and "Wellhausen, Heidentum, 2d ed., p. 30.
'^ Cf. his Arabia Deserta, Vol. II, pp. 511, 515, and 517.
s See CIS., Ft. II, Vol. I, Nos. 182, 183, 185, and De Vogiie's Syria
centrals, Vol. I, pp. 107 and 119.

« CIS., Ft. II, Vol. I, No. 190, etc. ^ Panarion, LI.

283

234 SEMITIC ORIGINS

in Edom, and tells us that the heathen Arabs at that place
drew a parallel between her and her son Dhu-'l-Shara on
the one hand, and the Virgin Mary and the child Jesus
on the other. 1 It appears, then, that Dhu-'l-Shara was
her son, and that she was an unmarried goddess. Robert-
son Smith 2 is no doubt right in interpreting this to mean
that she was originally a goddess of unwedded love, for
an unmarried virgin goddess was an unheard-of anomaly
among the ancient Semites.

Perhaps it was from Petra that some other ancient
writers heard of this goddess. Thus Herodotus speaks of
her^ under the name Alilat, and calls her son Dionysos.
Ephraem Syrus speaks of her and her companion goddess
Al-Uzza, and tells how women sacrificed chastity in their
honor.* Jerome also bears testimony to the same fact,
and tells us further that the goddess was identified with
the morning star.^

At Hegra she was also worshipped, but there the name
of Dhu-'l-Shara was placed before that of Al-lat.® At the
date from which the inscriptions from Hegra come the
influence of the patriarchal form of society had been felt
to such an extent that Dhu-'l-Shara had become superior
to his mother ; perhaps he had become her husband.

At Palmyra, in the second century a.d., the worship of
Al-Lat was coupled with that of the god Shamash.'' It is

1 Wellhausen {Heidentum, 2d ed., pp. 48, 49) seeks to break the force
of this because Epiphaiiius says the goddess was called Qaaba (xaa/SSu).
Wellhausen thinks the god was regarded as the offspring of the stone which
represented him. Robertson Smith is, I think, right {Religion of the
Semites, 2d ed., p. 56, n.) in giving the interpretation which I have adopted
in the text. Semitic gods were frequently so identified with the object
which represented them, that Epiphanius, no doubt, has put the name of
the stone fetich which represented her for the goddess herself.

2 See preceding note.
8 Book III, 8.

* Opera, Vol. II, pp. 457 E ; 458, 1. 1 ; 459 C.
6 Cf. Jerome's Vita Hilarionis, c. 25.
6 CIS., Pt. II, Vol. I, No. 198.
' De Vogue, Syria centrale, No. 8.

SURVIVALS 235

probable that the worship of Shamash had at this north-
eastern Nabathsean outpost been introduced from Baby-
lonia, and that it had been united with the Arabian cult
by the marriage of Shamash and Al-Lat.

The connection of the Al-Lat cult with the primitive
Semitic goddess is obvious. Al-Lat is but an epithet ^ and
was applied to the goddess at various points until it super-
seded her real name. All the features of her worship of
which we know are best accounted for in this way.^

Another Arabian goddess, who has been in recent years
proven to be a survivor of the primitive Semitic cult, is
Al-Uzza.^ She was, as we learn from the quotation which
Yaqut,* the Arabian geographer, makes from Ibn-al-Kalbi,
especially worshipped by the Koraish, the prophet's tribe,
whose headquarters were at Mecca. They honored her,
he says, with sacrifices and pilgrimages. In another pas-
sage he says that the place where her victims were slaugh-
tered was called the Ghabghab,* a name which seems to
have been applied to a rivulet or trench, into which the
blood of the victims drained,* and which emptied into the
Zemzem. The latter was a well which seems to have been
especially connected with her worship ; into it images of
sacred animals, such as the gazelle, which were offered in
her worship or in that of Allah, with whom she was con-
nected, were thrown.'^ In a similar way she, with Allah,
was connected with the Qa'aba, into which her golden
gazelles were afterward put.^ This connection with Allah
and the Qa'aba is established by the Qur'an, which makes
her one of Allah's daughters.^

1 The name seems to have been originally Al-Lahat, "the goddess,"
corresponding to Al-Lah ; of. Wellhausen, Seidentum, 2d ed., p. 33.

2 Cf. Hebraica, Vol. X, pp. 58-66.

8 See W. R. Smith's Kinship, pp. 294, 295, and Hebraica, Vol. X,
pp. 58-59.

* Cf. ed. Wustenfeld, Vol. Ill, p. 664. ^ Op. cit. Vol. Ill, p. 773.

8 Cf. Wellhausen, Heidentum, 2d ed., p. 103, Smith, Religion of the
Semites, 2d ed., pp. 198, 228, and 340.

' Ibn Hisham, Vol. I, pp. 93, 94.

8 Ibn Hisham, Vol. I, p. 94. ' Sura, 53».

236 SEMITIC ORIGINS

The connection of Al-Uzza with the Zemzem shows that
before the time of the prophet she had been a goddess of
wells or a ha'alat, and consequently a goddess of the soil
and of fertility like the old mother Athtar. The fact that
the dove was sacred to her, together with the nature of
her festivals, connects her worship with that of the Ashtart
of Phoenicia and its colonies, to whom the same bird was
sacred. 1

The most decisive indication of the direct descent of Al-
Uzza from the old mother goddess is the character of the
festivals celebrated in her honor. Isaac of Antioch testi-
fies that these feasts were licentioixs,^ that boys and maidens
were sacrificed in them,^ and that the goddess was identi-
fied with the planet Venus. This festival still survives at
Mecca. It is celebrated in the sixth month and is still of
a licentious character.* It is the lineal descendant of one
of the festivals of the primitive goddess described above
in Chapter III.^

At Nakhla, a valley southwestward from Mecca, which
takes its name from its abundant palm trees, Al-Uzza was
identified with a samura tree or group of samura trees,^
which, as noted above, are declared in a scholion to Ibn
Hisham to be palm trees. '^ The doubts of Wellhausen
and Robertson Smith as to the correctness of this state-
ment have already been discussed. The general course of
the development of Semitic civilization and the agencies
which acted as factors of progress tend, as we have traced
them, to establish the veracity of this scholion. Some
Arabic writers declare that there was a temple of Al-Uzza
at Nakhla, but Wellhausen^ is probably right in holding
that the temple was at another place called Bass, and that
the later Arabs confused this temple with Nakhla, where

1 Smith, Kinship, p. 294.

2 Ed. of Bickell, p. 244. Cf. Wellhausen, op. cit., p. 39 ff.
8 Ed. of Biokell, p. 220.

* See Snouck Hurgronje's Mekka, Vol. II, pp. 59-61. ^ p. 94 ff.

' Wellhausen, Heidentum, 2d ed., p. 38.

' p. 145. Cf. above, p. 77 f£. s Op. cit, p. 38.

SURVIVALS 237

the goddess was supposed to dwell in the trees already
mentioned.

It was probably a sacrifice in worship of Al-Uzza which
Theodulus, son of Nilus, witnessed " to the morning star "
among the Arabs of the Sinaitic peninsula. None of this
sacrifice could remain till the morning.^ The ritual of
this offering resembles that which Bent found in Abys-
sinia, which has already been traced to the primitive
Semitic cult.^

Wellhausen and Robertson Smith ^ have perceived that
Al-Lat and Al-Uzza are in reality one, and that their
names are but epithets for the same goddess. As Al-Lat
is the feminine of Al-Lahu, so Al-Uzza, " the mighty," is
an epithet applied in its masculine form to Allah also, and
in south Arabia was applied to other deities.* There can
be no doubt that both are survivals in slightly different
forms of the primitive mother goddess of the Semites, who
was in part transformed at Mecca into, and became the
basis of, the Mohammedan Allah.

With the movement of the Semites northward from
Arabia the worship of their mother goddess was, as we
have seen, carried. In many places she was in prehistoric
times transformed into a god ; but in others she survived
in her original character far down into historic times.

One of the places where such survival occurred was a
city which occupied an important site on the plateau east
of the Jordan, and which took its name from the goddess.
It appears in the earliest extant documents which refer to
that section of the country, the inscriptions of Thothmes
III^ and the El-Amarna® letters. It is called in these

iSee Migne, Patrologia Grmca, Vol. LXXIX {Nili Opera), p. 611 ft.
Cf. Wellhausen, op. cit., p. 42 ff., and Smith, Seligion of the Semites,
2d ed., pp. 166, 227, 281, 338, 361, 363, 364.

2 Above, p. 112. ^ Heidentum, 2d ed., p. 44 ; Kinship, p. 295.

* See the name Il-'Azza, " God is mighty," whioli occurs as a Sabsan
proper name in CIS., Ft. IV, Vol. I, No. 118.

6 Cf. W. Max MliUer's Asien und Ewropa, p. 162.

6 Cf. KB., Vol. V, Nos. 1421" and 23721.

238 SEMITIC ORIGINS

documents Ashtart, and in the Old Testament Ashtaroth ^
and Ashtoreth Karnaim.^ It was therefore a prehistoric
sanctuary of the goddess. It continued to be an important
centre of her worship down to the Maccabaean period.^

What the name Karnaim (" two-horned ") signified has
been a matter about which opinions have differed, some
taking it to mean that she was a moon goddess,* others
that she was worshipped in the form of a cow.^ The real
meaning is probably that suggested by Moore,® which
makes Karnaim " the two-peaked mountain," and supposes
that the city was situated in a valley between two hills.
In later times the name was shortened to Karnaim," and
under this name maintained its existence down to the
second century^ and perhaps later. By the time of Euse-
bius its importance had apparently waned.^ Although we
have no details concerning the worship of the goddess in
this city further than that she had a temple there, there
is no reason to suppose that it differed in any material
feature from the forms which it assumed elsewhere.
The writer of 2 Maccabees has confused the Ashtart of
Karnaim with the goddess Atargatis, a divinity whose

1 Josh. 13.21 = Gen. UK

'The Onomastioa of Eusetoius and Jerome give two places east of
the Jordan named Astarte. Buhl, in his Geographie, p. 248 ff., follows
this, but it is doubtful whether it is true. The names floated about some-
what in the later times (cf. G. A. Smith's "Ashtaroth" in Encyc. Bib.,
and my "Ashtaroth" and "Ashtoreth Karnaim" in Jewish Encyclo-
pedia). Tell Ashtarah, Tell Ash'ari, and Muzeirib are sites which have
been identified with the name. Excavation of the sites will be necessary
before their identity can be determined.

* See Stade in ZAW., Vol. VI, p. 323 ff.

5 That was my view in 1894. Cf. Hebraica, Vol. X, p. 40.

^ JBL., Vol. XVI, p. 156 ff. Moore bases his suggestion on a parallel
name of Baal contained in some Latin inscriptions from North Africa.

' So Amos 613, 1 Mac. 5«, and 2 Mac. 122s. Cf. Wellhausen, Sldzeen
und Vorarbeiten ; Heft V, p. 86; G. A. Smith, Book of the Twelve
Prophets, Vol. I, p. 176, and Nowack, Kleinen Propheten, p. 147.

8 1 Mac. 5", and 2 Mac. 1226.

' This is inferred in consequence of the probable confusion of names
by Eusebius.

SURVIVALS 239

origin and nature are much debated. Baethgen has
pointed out^ that the name Athtar or Ashtar would in
Aramaic become Atar. Such a goddess was found by
Assurbanipal among the Aramaeans, who were associated
with the Nabathaeans encountered in his Arabian cam-
paign. ^ Her worshippers called her Atar-samain, i.e.
"Atar of the heavens." This proves the presence among
the Aramaeans of the old Semitic mother goddess. A god-
dess who is in part at least the same appears in a bilingual
inscription (Aramaic and Greek) from Palmyra under the
name 'Atar 'atah, in Greek Atargatis.^ Atargatis is also
the name of a goddess who is mentioned by Greek and
Roman writers.*

Since the second element of the name of the goddess in
the Aramaic portion of the Palmy rene inscription appears
as a component element of theophorous proper names,^
Baethgen concludes that it, too, was originally the name of
a deity. Since Lucian and Macrobius describe the temple
and rites of Atargatis at Hierapolis-Bambyce (Mabug) in
Syria, where she was worshipped as the consort of the
god Hadad,^ and since Melito of Sardis and some Greek
inscriptions from Batanea couple a goddess 'Ati with
Hadad, Baethgen also concludes that Atargatis is a name
compounded of the Aramaean goddess Atar and that of
the goddess 'Ati, and formed like the name Ashtar-

' Beitrage zur semitischen Seligionsgeschichte, p. 69 fi. The s or 8
would in Aramaic become t, and the two fs would be assimilated or
written with a dagesh understood.

2 Cf. Ill R., 24, 11. 98 and 106, and George Smith's Assurbanipal, pp.
270, 271, 283, and 295.

' De Vogile, Syrie centrale, No. 3. The name is spelled iny-inji on a
coin. Cf. ZDM&., Vol. VI, p. 472 ff.

*Cf. Lucian, De Syria Dea, §§ 14, 15; Strabo, Book XVI, 1, 27,
and Macrobius, Saturnalia., I, 22, 18.

' In addition to the names cited by Baethgen, op. eit. , cf . Cook, Glos-
sary of Aramaic Inscriptions, p. 95 ; Lidzbarski, Nordsemitische Epigra-
phik. Vol. I, p. 347, and Gottheil in JAOS., Vol. XXI, Pt. II, pp. 109-111.

sin addition to the references in n. 4, cf. Lucian, op. cit., §§ 31, 32,
and Pliny, Nat. Hist., V, 23.

240 SEMITIC ORIGINS

Chemosh. The Phrygians had a goddess Attis,i and
since the name Atargatis occurs only in late sources, it is
possible that a fusion of two goddesses, one Semitic and
one foreign, had taken place. Jensen,^ on a far more
slender thread of evidence, explains the name as a corrup-
tion of that of the Hittite god or goddess Tarkhu, with
an 'Ayin prefixed and a feminine ending added to make it
analogous to Ashtart. Such an explanation seems far less
probable than Baethgen's.

Still another possibility should be considered. 'Ati or
'Athi, the second element in the name of the Palmyrene
goddess, may originally have been an epithet descriptive
of her as the defender of her people, and finally by a
fashion similar to that which attached the name Sebaoth
to the name Yahwe among certain Israelitish writers it
may have become a part of the name of the goddess. If
it were an epithet, it would naturally come in time to be
used for the goddess herself, and thus would enter as an
element into proper names as Ahu, Akhu, and Meleh have
entered. This explanation cannot be regarded as very
satisfactory, as no good Semitic etymology of 'Ati is
forthcoming.

On the whole the explanation of Baethgen seems most
probable. The late date of the sources in which the name
is found, the well-recognized syncretism which took place
in Syria after the days of Alexander the Great, as well
as the tendency of the Semites of the Greek and Persian
periods to form divine names by compounding the names
of two separate gods,^ all point in this direction. If such
combination took place, it was after the time of Assurbani-
pal's campaign (cir. 640 B.C.), — and probably long after,

1 Cf. Macrobius, Saturnalia, I, 21, 7, and 22, 4 ff., and Lucian, op. cit.,
§ 15. Attis and Atargatis are tiotli said by these writers to ride upon
lions — another point in favor of the theory of fusion.

2 Hittiter unci Armenier, p. 157.

2 See the article on " West Semitic Deities with Compound Names "
in JBL., Vol. XX, p. 22 ff.

SURVIVALS 241

— for the Semitic goddess in her Aramaic form at that
time was worshipped under her own name.

Atar, even when compounded with the Phrygian god-
dess, did not differ materially from Ashtart in character
and functions, as Lucian and Macrobius testify. We are
therefore justified in regarding her as practically the same
as the Aramaean goddess Atar. Atar was as like to Ash-
tart as was Ashtart to Ishtar. All had sprung from the
same root, but had developed in different branches of the
great Semitic family. It is not strange therefore that
the author of 2 Maccabees should identify the two and call
the temple of Ashtart the temple of Atargatis. Consider-
ing the tendency to fusion in the later time, this seems
more probable than the supposition that there was in
Karnaim a temple of both goddesses.^ The writer prob-
ably called the goddess by the name most familiar to him.

The Aramaic goddess Atar (Atargatis) was, like her
sister goddesses in other countries, worshipped sometimes
alone as among the Isammikhi ^ and at Palmyra, and some-
times as the consort of the closely related deity Hadad.
At Palmyra, where the population was composed in part
of Nabathsean Arabs and of Aramaeans, both Al-Lat and
Atargatis found worshippers. At Hierapolis-Bambyce
she was worshipped in the temple of Hadad as his consort.
Fishes and doves were sacred to her. Statues in the
temple represented both her and Hadad, that of the god
being supported by bulls, and that of the goddess by lions. ^

Another city in which the worship of this goddess sur-
vived and where its history seems to have run a similar
course was Ashkelon. This town, situated on or near the
Mediterranean coast, was a fortress of some importance
under the eighteenth Egyptian dynasty ; * it joined in the

1 So Cheyne, Encyc. Bib., Vol. I, col. 379.

2 III R., 24, 11. 98, 106, and George Smith's Assurbanipal, pp. 270,
271, 283, and 295.

3 See Luoian, De Syria Dea, §§ 14, 15, 31, 32, and Macrobius, Satur-
nalia, I, 22, 17-22.

* Cf. KB., Vol. V, No. 211 ff.

242 SEMITIC ORIGINS

conspiracy against Jerusalem -which Abdi-kheba's letters
reflect;^ and seems to have revolted from Rameses IP.
In the early days of Israelitish history it was one of the
five Philistine cities of importance.

The earliest mention which we have of the temple of
the goddess at this place is probably the statement in
1 Sam. 31^, that the Philistines after the battle of Gilboa
hung the armor of Saul in the "house of the Ashtarotb."
As we have positive evidence afterward of the worship of
the goddess only at Ashkelon, the reference is probably to
the temple of that place.' If this be true, the goddess of
Ashkelon was to the Israelites indistinguishable from the
Canaanitish goddesses.

We catch a glimpse of her cult again in Herodotus,*
who calls her the Oriental Aphrodite, and who by refer-
ence to the disease which the Scythians took from thence
bears witness to the survival of those rites which we have
found to be so characteristic of the primitive Semitic
mother goddess.

Later writers^ call the name of the goddess Atargatis,
identifying her with the Aramaic-Syrian deity which we
have already traced. It is impossible from our present
information to determine absolutely whether it was the
Aramaeans or the Canaanites who first planted the worship
of the goddess at Ashkelon. It seems reasonable to con-
jecture, however, that it was of Canaanitish origin, and that
the Aramaic element was afterward introduced into it.

It was probably at Ashkelon that the custom of repre-
senting the goddess as half woman and half fish originated,
for at Ashkelon her temple stood near a lake filled with

'^KB., Vol. V, No. 180" ff.

^ W. Max MuUer's Asien und Europa, p. 222.

^ Of course it is possible that there may have been smaller temples of
the goddess elsewhere, but on the whole the position taken in the text
seems probable.

* Book I, 105.

^ See Diodorus Siculus, Book II, ch. iv, and Ovid, Metamorphoses,
Book IV, 11. 44-46.

SURVIVALS 243

fish,^ and was also not far from the sea-coast. Fishing
must have become one of the means of living at a very
early time, and by the same processes of thought which
led to the representation of Ea at Eridu as half man,
half fish, the goddess of the fishermen of Ashkelon took on
a similar form. At Hierapolis-Bambyce, situated between
the ranges of Lebanon, she was represented under a similar
form, but the custom may have been transferred thither
from Ashkelon. It has been inferred ^ from this form that
the goddess was the personification of the fructifying
power of water. It seems clear from the wide survey of
the cult which has been made in the preceding pages that
such connection of the goddess with water far antedates
the time when her worship was planted at Ashkelon.
The fish form, though, may well have been born in that
peculiar environment.

By Greek writers Atargatis was more often called
Dekerto. The myth, that becoming enamoured of one
of her worshippers, she became by him the mother of
Semiramis, the queen of Babylon,^ is additional proof of
her practical identity with the old polyandrous goddess
which was worshipped among the Nabathseans as Al-lat
and sometimes held in later centuries to be a virgin.

At Sidon the chief deity was, as already noted, Ashtart,
but in this city her worship underwent a twofold develop-
ment. On the one hand, the old mother goddess was re-
tained in her primitive independence, so that when her
cult was opposed by Israelitish prophets she was called
"the abomination of the Sidonians."* To her, king Tab-
nith tells us, both himself and his father were priests,^
and Eshmunazer II says that his mother was her priestess.*

^ Diodorus Siculus, ibid.

2 Legarde, Mittheilungen, Vol. I, p. 77, and White's article, " Atarga^
tis" In Hastings's Dictionary of the- Bible.
" Diodorus Siculus, Book II, ch. iv.
« 2 Kgs. 2.318.

' Cf. Bevue archeologique, 1887, p. 2.
» CIS., Vol. I, Pt. I, No. S"!".

244 SEMITIC ORIGINS

Side by side with this primitive cult there was one
somewhat modified. "Ashtart of the name of Baal," an
Ashtart probably which had in part been metamorphosed
into a god, was also worshipped. To each of these god-
desses Eshmunazer built a temple,^ perhaps the same
which Lucian afterward saw there. ^

Sidon was one of the headquarters of the Phoenician
shipping trade, and its goddess became in consequence
the patroness of mariners. She is often pictured on
Sidonian coins as standing on the prow of a galley with
one hand outstretched, holding a crown and pointing the
ship on its way,^ a device also adopted on the coins of
other Phoenician cities. According to Lucian, Ashtart
of Sidon was also identified with the moon.*

It is stated by Eusebius on the authority of a Phoenician
writer that at Tyre, Ashtart was the chief deity with
whom two others (probably Melqart and Eshmun) ^ were
associated, and that the goddess here had the head of a
bull. Josephus also states that Hiram, king of Tyre,
built, at Tyre, in addition to the temple of Baal, a temple
of Ashtart, in which Eth-Baal, the father of Jezebel, was
priest. Josephus no doubt gives the correct view in say-
ing that Melqart (Heracles) was at the head of the
pantheon of Tyre, and that the worship of Ashtart was
in historical times subordinate to his.

At Byblos, the ancient Gebal (the Gubla of the El-
Amarna letters),^ the cult of Ashtart survived in much
of its original form. Yahumelek,'^ a king of Gebal in the

1 CIS., Pt. I, Vol. I, No. 3.

2 De Syria Dea, § 4.

' Cf. Driver in Hastings's Dictionary of the Bible, Vol. I, p. 167.

* Op. cit., § 4.

6Cf. my article "The Pantheon of Tyre," in JAOS., Vol. XXII,
p. 11.5 ff., and Herod., II, 49, who vouches for the Tyrian origin of Aphro-
dite at Thebes, with whom Adonis was connected. Cf. Pans., IX, 16, 3,
and below, Chapter VIII.

« See KB., Vol. V, Nos. 50, 53, 123, and 137.

' CIS., Pt. I, Vol. I, No. 1.

SURVIVALS 245

fifth century B.C., venerates the goddess as the Bdalat of
Gebal, and makes it clear that she not only stood at the
head of the pantheon, but that all other worship there was
practically subordinate to hers. Her temple at Byblos is
pictured on an old coin,^ and is also mentioned by Lucian.
It was here that Lucian found the rites of Tammuz surviv-
ing in connection with the worship of the mother goddess.
He describes it as follows : —

" But I also saw in Byblos a great temple of Aphrodite
of Byblos, in which also the rites to Adonis are performed.
I also made inquiry concerning the rites ; for they tell the
deed which was done to Adonis by a boar in their own
country, and in memory of his suffering they beat their
breasts each year, and waU and celebrate these rites, and
institute great lamentation throughout the country. But
when they have bewailed and lamented, first they perform
funeral rites to Adonis as if he were dead, but afterward
upon another day they say he lives, and they cast (dust)
into the air and shave their heads as the Egyptians do
when Apis dies. But women such as do not wish to be
shaven pay the following penalty : On a certain day they
stand for prostitution at the proper time ; and the market
is open to strangers only, and the pay goes as a sacrifice
to Aphrodite."^ . . .

"But there is also another marvel in the country of
Byblos : a river from Mount Libanos empties into the sea.
The name of the river is Adonis. But the river each year
becomes bloody, and having lost its own complexion, falls
into the sea and reddens a large part of the sea, and
gives the signal for the lamentations to the inhabitants of
Byblos. They say that in these days Adonis is wounded
on Libanos, and his blood going into the water changes
the river, and gives to the stream its name. The majority
tell this. But a certain man of Byblos, who seemed to

1 See Pietschmann, ffeschichte der Phmnizier, p. 200, and Journal of
Bellenic Studies, Vol. IX, p. 215.

2 Lucian, op. cit., § 6.

246 SEMITIC ORIGINS

me to tell the truth, adduced another cause of the suffer-
ing. He spoke as follows : ' The river Adonis, O stranger,
comes through Libanos ; but Libanos has a great deal of
yellow soil. Therefore, the hard winds in these days set-
ting upon the soil bear it into the river — the soil being
of an especially red color ; and the soil gives it its bloody
tint ; and the country is the cause of this suffering, and
not the blood as they say.' The Byblite adduced such
causes to me, and if he related these things to me accu-
rately, the incident of the wind seems to me especially
supernatural."^

It is clear from these passages that the myth of the
death of the son of the old mother goddess survived at
Gebal in much of its primitive form. True, the myth had
taken on a local coloring, and connected itself with local
circumstances; but the rites attached to the celebration of
the god's resurrection are in many respects still primitive.
Some progress has been made since women coiild be shorn
in lieu of a more degrading sacrifice, if they desired, but
this progress does not hide the features in the rites which
have survived from the Semitic matriarchal past. The
worship of Ashtart was no doubt prevalent at many points
in Canaan, whence it was adopted at various times by the
Israelites, but we have not now the means of tracing it
in detail.^

Another goddess which sprang from the same root as
those we have been considering, and which exercised the
same functions, though called by a slightly different
name, was the goddess Ashera. In the period repre-
sented by the El-Amarna tablets she was apparently the
goddess of a tribe called the Bne-Ebed-Ashera.^ From a
Sumerian hymn published by Reisner * we learn that she

1 Lucian, op. cit., § 8.

2 Cf. Jud. 213, 106, 1 Sara. 7*, Jer. V«, and Eze. 8".

' See KB., Vol. V, No. 53 ff., and cf. JBL., Vol. X, p. 82 H.

* See Mittheilungen d. kgl. Museen zu Berlin, Heft X, p. 92 ; also a
hematite seal in ZA., Vol. VI, p. 161, an astronomical text in ZA., Vol.
VI, p. 241, and the remarks of .Jensen, ZA., Vol. XI, p. 302 ff.

SURVIVALS 247

was the consort of the god of the Westland, i.e., Hadad.
The tablet from which the hymn is published dates from
the Greek period. We do not know that the goddess was
worshipped alone in the fifteenth century, though it seems
probable that she was. If she was then an independent
mother goddess, she might long before the third century
have become in some localities the consort of Hadad. In
this latter character she was practically identical with
Atargatis. From this fact it follows that from the first
she must have been the old mother goddess under another
name. She was probably known among the Aramaeans
much earlier than this, for she seems to have been known
to Khammurabi,! king of Babylon about 2300 B.C. The
name of this goddess also appears according to our present
text in three passages of the Old Testament,^ but it is
thought that in every case the text has been corrupted or
glossed,^ and that the original reading was Ashtoreth
(Ash tart).

Hommel has pointed out that Athirat, the Arabic
equivalent of Ashera, appears as the consort of the god
Wadd in a Minsean inscription.* Since the worship of
the god Hadad found its way into Arabia, both as Hadad
and as Ramman,^ it may be that the worship of Ashera
found its way thither from Syria in the same way. This
is not probable, however, since she appears as the consort
of the native Arabian god Wadd. Athirat is therefore to
be regarded as a native Arabian product, brought forth
by forces analogous to those which produced her Syrian
counterpart.

It is a well-known fact that Ashera was in the Old

1 See King's Letters and Inscriptions of Hamrrmrabi, No. 66, and Hom-
mel's Aufsiitze iind Abhandlungen, Vol. II, p. 211 it.

2 Of. Jud. 3', 1 Kgs. 181S, and 2 Kgs. 23*.

8 Cf. Moore's article "Asherah," §2, in Encyc. Bib., and Budde in
New World, Vol. VIII, p. 734.

* Cf. Expository Times, Vol. XI, p. 190, and Aufsdtze und Abhand-
lungen, Vol. II, p. 206 ff.

' See above, p. 227.

248 SEMITIC ORIGINS

Testament period a post or pole which was planted by the
altars of the different gods,^ which was sometimes carved
into revolting shapes,^ and probably sometimes draped.^
G. Hoffmann has shown* that these posts originally
marked the limits of the sacred precincts of the shrine,
and in the Ma' sub inscription the name is equivalent to
" sacred enclosure." Moore finds ^ in this the explanation
of the use of the word in Assyrian {asMrtu, ashrdti,
esMrtu, eshrdti), in the sense of sanctuary.

It is probable that the application of the name to the
goddess arose from the connection of these points with
her sanctuary. If one or more of these were carved into
a rude representation of the goddess, it would be very
natural for the name to pass from the post to the deity.

This seems to have happened independently in three
centres, — in Arabia, in Syria, and in Assyria.^ This pro-
cess did not in any way change the nature or the functions
of the deity, but simply gave her a new name. As pointed
out above, Ashera is in the El-Amarna tablets the goddess
of a tribe, and, it may be added, of a sheik who was the
head of a tribe. The suggestion made in 1895 ^ that the
Bne-Ebed- Ashera is the same clan, or the nucleus of it,
which appears in the Old Testament as Asher, still seems
most probable. The Egyptian monuments show that
under Seti and Rameses II of the nineteenth Egyptian
dynasty this tribe was still in Palestine.^ The Israelitish
traditions classed it with the children of Jacob's concu-
bines, showing that they had a consciousness that it was
among the latest to join the Israelitish confederacy. It is
therefore probably a tribe of Aramaic extraction, which
became amalgamated with the Israelites after their settle-
ment in Canaan.

■^ Cf. Moore's article "Ashera" in Encyc. Bib., and mine on same
Bubject in Jewish Encyclopedia.

2 1 Kgs. 1512. 5 Encyc. Bib., as above,

a 2 Kgs. 28'. 6 See above, p. 223.

* Ueber einige phoen. Inschriften, p. 26 ff. ' JBL., Vol. XV, p. 174.
s See W. Max Miiller's Asien und Europa, p. 230.

SUBVrVALS 249

In all probability the goddess of the tribe became a god
soon after the El-Amarna period, for the name appears on
the Egyptian monuments in its masculine form, as it does
in the tribal name in the Old Testament. As the Israel-
itish nation was welded into a confederacy, the various
tribal gods were either identified with Yahwe or banished.
While the latter seems to have been the fate of the god of
the tribe of Gad,i the former was the fortune of the god
of the tribe of Asher. This appears from the fact to
which Hommel has called attention,^ that in Deut. 33^
Asher is an alternative name of Yahwe. ^

As the Phcenicians in their restless movements for
trade and colonization progressed westward, they carried
the worship of the goddess Ashtart with them, and scat-
tered it all over the islands and shores of the Mediter-
ranean. In many of the localities where it was thus
planted it can afterward be traced. For the present
we shall confine our attention to localities where the
Semitic element continued to be tolerably distinct. There
are several of the islands of the Mediterranean where this
was the case. The sources of our information are Phoe-
nician inscriptions, Greek inscriptions, and Greek and
Roman writers. In the Greek sources the goddess is
usually called Aphrodite, and in the Latin, Venus; but
there can be no doubt of the identity of the divinity of
whom they speak.

In the island of Cyprus, which lies nearest to the coast
of Phoenicia, this Semitic worship was naturally planted
at a very early time — how early we cannot tell. In the
Homeric poems Aphrodite is already spoken of as Cyprian,*
and her temple at Paphos is referred to.^ It was then no

1 See Isa. 651^, and cf. Oriental Studies of the Oriental Olub of
Philadelphia, p. 108.

" Aufsdtz und Abhandlungen, Vol. II, p. 209.

3 The passage should be translated, "Happy art thou, Israel, a
people saved by Yahwe, the shield of thy help, and Asher, the sword of
thy excellency."

■• Iliad, V, 330. ^ Odyssey, VIII, 362 ff.

250 SEMITIC ORIGINS

doubt very old. Tradition assigned its foundation to one
Cinyras,^ who plays a considerable part in Cyprian my-
thology. The priests of the Paphian shrine were after-
ward supposed to be his descendants and bore his name.^
Of the early history of this worship we have no real data.
These Greek legends and myths can hardly be historicaL
A number of the German Assyriologists believe that the
letters of the king of Alashia^ to the king of Egypt,
which were found in the El-Amarna correspondence, are
really letters from Cyprus ; but even if they are, they make
no mention of religious matters, and so leave us as much
in the dark with reference to the religious status of the
island in the fifteenth century B.C. as though we did not
possess them. The Greek inscriptions written in the
Cypriotic syllabary testify to the existence of the goddess
at Paphos, but do little more than that.* Monuments have
been recovered which were dedicated to the goddess at
Paphos on behalf of various Ptolemies from 164-88 e.g.,'
as well as on behalf of the Roman Emperor Tiberius.*
These attest that the worship was flourishing during those
centuries. From Strabo'^ and Pausanias^ we learn that
the shrine at Paphos was still important in their days,
while Johannes Lydus * in the sixth century a.d. implies
that the worship had then ceased.

At Kition, in the southeastern part of the island, traces of
a temple of Ash tart also appear, i" We lack, however, the
means of tracing its history. One fragmentary inscrip-

1 Iliad, XI, 19-23, and Tacitus, Hist., II, 2, 3.

= Tacitus, Hist., II, 2, 3.

» Cf. KB., Vol. V, Nos. 25-32.

* Cf. Collitz, Sammlung der griechischen Dialekt-Inschriften, Gottin-
gen, 1884. Vol. I, p. 13, No. 1.

5 Cf. Journal of Hellenic Studies, Vol. IX, pp. 229-231, No. 14 ; p.
232 ff., No. 21 ; p. 233 ff.. No. 24 ; p. 240, No. 50.

' Journal of Hellenic Studies, Vol. IX, p. 227, No. 6.

' XIV, 6, 3 (683).

8 VIII, 5, 2.

' Be Mensibus, IV, 45.

w Cf. cm, Pt. I, Vol. I, Nos. 11 and 86.

SURVIVALS 251

tion ^ reveals the fact that a large temple retinue, consist-
ing of sacred prostitutes ^ or priests, slaughterers, barbers,
and slaves were maintained. These facts vouch for the
identity of the worship at Kition with other phases of the
primitive Semitic cult.

The temple of Ashtart at Paphos has been excavated
and its form may be studied in considerable detail.^ It
was evidently a Semitic temple, built on the same general
plan as the temple of Solomon at Jerusalem, but with con-
siderable variation in details. It was more than once in
later times destroyed by earthquakes, and rebuilt by the
Romans.* In the temple there was no statue of the god-
dess, but she was represented by an old Semitic masseba.^
Doves were sacred to her^ and many images of them
have been found in her temple. She was regarded as a
mother goddess, and was addressed as " mother." The
Semitic feast of the old mother goddess was kept to her in
the springtime, when a lamb or sheep was sacrificed to
her.' Only male victims were sacrificed to her, and kids
were regarded as the best for the purposes of divination,

1 CIS., Pt. I, Vol. I, No. 86.

* It is probably thus that the term, D'sbs, "dogs," should be inter-
preted. The term occurs in Deut. 2.3"- 1', where it seems to mean " male
priestly prostitute " (cf. Driver's Deuteronomy, p. 264 ff., and Steuerna-
gle's Deuteronomium und Joshua, p. 86 ff.). Clement of Alexandria so
understood the term and rendered it "fornicator" {Paidagogos, III, 3).
One consecrated to a god was perhaps so called because of his fidelity
in following his god (cf. W. R. Smith, Beligion of the Semites, 2d ed., p.
292). We have a Biblical instance in Caleb, i.e., "the dog who followed
Tahwe" in Num. 321^. This usage probably extended to Babylonia, for
the real names of the kings of Shirpurla, commonly called Ur-Nina and Ur-
Bau, were probably Kalbi-Nina and Kalbi-Bau (cf. Radau, Early Baby-
lonian History, p. 144) i.e. "Dog of Nina" and "Dog of Bau."

8 Cf. Journal of Hellenic Studies, Vol. IX, pp. 19.3-215.

* Cf . Journal of Hellenic Studies, Vol. IX, p. 193.

« Tacitus, Hist., II, 3; Serv. Aen., I, 720. Cf. Hehraica, Vol. X,
p. 46 ff.

« Antiphanes, ap. Athen., VI, 71, p. 257 ; XIV, 70, p. 655, and the
"PaphisB columbsB" of Martial (VIII, 28).

' Johannes Lydus, De Mensihus, 45.

262 SEMITIC ORIGINS

in whicli her priests were thought to be especially skilf ul.i
No blood was shed upon her altar, and though the masseba
stood in the open air it was thought that it was never
rained upon.^ The devotees of the goddess were initiated
by impure rites,^ and parents often dedicated their children
to the goddess.* In later times there was much admixture
of Greek elements into the Paphian worship, but never-
theless the Semitic type of goddess on the whole pre-
vailed.^ It was from Cyprus as the Greeks themselves
believed that the worship of Aphrodite spread to the
islands and coast lands of Greece.

In Crete the worship of this goddess was also estab-
lished at an early time, and the Cretans themselves be-
lieved that their island was the original home of the cult.®
In the island of Rhodes she was worshipped along with
Apollo and Jisculapius,^ who were no doubt originally
Baal and Eshmun,^ but who through Greek influence were
transformed into Greek gods. Her worship was also
planted in the island of Malta by a Phcenician colony,*
though the traces of it which remain are slight. A very
important and ancient seat of it was at Eryx in the island
of Sicily, whence its influence spread through that island
to Carthage, and into many parts of Italy, extending

1 Tacitus, Sist., II, 3.

2 Tacitus, Hist., II, 3.

8 Clement of Alexandria, Protreptikos pros Hellenes,' pp. 12, 13 ;
Arnotiius, adv. Oentes, V, 19 ; Justin, XVIII, 5. Herodotus, after describ-
ing the impure rites of this goddess at Babylon (I, 199), adds, "In some
parts of Cyprus there is a custom very similar."

* Cf. Journal of Hellenic Studies, Vol. IX, p. 228, No. 8 ; p. 235, No.
33 ; p. 236, Nos. 35, 39 ; p. 237, Nos. 41, 42.

5 Cf. Dyer, 27i6 Gods of Greece, oh. vii, and Driver in Hastings's Dic-
tionary of the Bible, Vol. I, p. 170. That vegetation was thought to be
connected with the goddess in Cyprus as in ancient Arabia is shovni in
Ohnefalsch-Richter's Kypros, pp. 118-126.

s Cf. Diodorus Siculus, V, 77, and Farnell, Cults of the Greek States,
Vol. II, pp. 631-633.

' Cf. Bull, de Corr. Hell, 1880, p. 139.

8 See cm, Ft. I, Vol. I, No. 143 ; cf. also JAOS., Vol. XXP, p. 188 fl.

« Cf. CIS., Pt. I, Vol. I, No. 132.

SURVIVALS 253

especially to Rome.i In Sicily the goddess was as else-
where served by a troop of female priestesses,^ whose
character and functions we can from our previous knowl-
edge easily divine. Here the dove was also sacred to the
goddess, and there were two feasts, in reality parts of the
same festival, the dates of which were supposed to be con-
nected with the flight of the doves. ^ This cult may be
traced further among the islands and into the mainland of
Greece and Italy, but that task belongs rather to another
part of the work,* since the goddess there took on such a
foreign character. It is clear from the evidence already
cited that in the Phoenician colonies of the Mediterranean
islands all the essential features of the old Semitic mother
goddess were preserved. At each sanctuary a certain
local coloring was given to her myths as was natural and
as was the case in other places, nevertheless she remained
the unmarried mother goddess, fostering sexual love,
maintaining a retinue of priests and priestesses who kept
the atmosphere of social life impure bj^ perpetuating under
the guise of religion the long outgrown customs of a
barbarous civilization.

It has already been pointed out^ that the Semitic mother
goddess, whose ciilt we can trace through so many countries,
was also established in North Africa. There, in the period
from which our Punic inscriptions come, she seems to have
been in part subordinate to, and in part superior to, her
masculine counterpart Baal-Hamman. The name by which
she was known in Africa was Tanith, which was, perhaps,
given her as the one who increased life and blessings.^

1 Cf. Diodorus Siculus, IV, 83 ; Paus. VIII, 24, 6 ; Polybius, I, 55 ;
Strabo, VI, 2, 5 ; and Virg. Aen., I, 750.

2 Strabo, VI, 2, 5. * Chapter VIII.
' JElian, De Natura Anamalium, IV, 2. ^ Above, p. 150.

^ Georg Hofimann, Ueber einige Phoen. Inschr., p. 32, holds that the
name is a rebus, made from the final letters of mrH^U, IS)n"'?153 and mi-
He believes that the Greek Ai5u, AetSu came from a corrupt pronunciation,
in which the t's were hardened to d's and then J assimilated as in nO- This
is ingenious, but it is not convincing. I would, with all reserve, suggest

254 SEMITIC ORIGINS

The superiority of Tanith to Baal is shown in the fact
that in the votive inscriptions she is regularly addressed
first ; ^ her subordination to him, in the fact that officially
he seems to have been the head of their pantheon. ^ That
she held the chief place in the popular thought is shown
by the fact that no prayer for life and blessings seems to
have been complete which did not include an address to
her.^ As pointed out above, the goddess had at some
time passed in part through the process of transformation
from feminine to masculine, Baal-Hamman being in fact a
differentiation from her.* She was still represented by an
image, feminine in form but with a bearded face, and is,
therefore, addressed continually as " Tanith with the face
of Baal." 6

Tanith was a mother goddess, and upon her feast days
songs were sung and deeds were enacted which, according
to Augustine, shocked all modesty.® Georg Hoffmann
has pointed ouf^ that Dido is but another form of the
name Tanith, and the identification is accepted by others.'
It is no doubt correct to see in Dido another form of

that possibly the name is a noun of the form of the infinitive from the
stem [m (cf. Arab. , .jj. and Bth. watan). Such infinitives are formed
by the elision of the 1 and the addition of n, and are not uncommon in
all the great branches of Semitic speech (cf. Barth, Nominalbildung
in den semitischen Sprachen [1889], p. 122). The fourth stem of this root
means in Arabic "to multiply," "increase." The only place vfhere it
occurs in north Semitic so far as I have observed is in Phoenician, where
it appears naturally as ]tT, and is used as a synonym of ]n, "to give"
(of. the references in Bloch's Phoenicsiches Glossar, p. 33). If thus
derived, the name would mean "the giver," "multiplier," or "increaser,"
and would be most fitting for a goddess of fertility.

1 Cf. the hundreds of votive inscriptions from North Africa in CIS.,
Pt. I, Vol. I, No. 180 ff., and those published by Berger in Actes du
onziime congres international des orientalistes, Pt. IV, p. 273 fi.

2 Cf. CIS., Pt. I, Vol. I, Nos. 165, 167. » Ibid.
* Above, p. 150.

8 See the references in n. 1, and also JAOS., Vol. XXP, p. 187.

° De Civitate Dei, II, 4.

' See reference in p. 253, n. 6.

8 Cf. W. R. Smith, Religion of the Semites, 2d ed., p. 374.

SURVIVALS 255

Tanith, whether the name be the same, or whether it be
derived from another epithet. ^ This identification enables
us to see in the tale of Dido's love for ^neas ^ told by
Virgil, another evidence of the survival in Tanith of the
peculiar characteristics of love which were embodied in
the old Semitic deity. Indeed it is probable, as Farnell
has pointed out,^ that the whole ^neas story is but a
translation into poetry of the myths of this cult.

In later times, as other Phoenicians came into North
Africa, they brought with them anew the worship of Ash-
tart, so that at times Ashtart and Tanith appear side by side
as different goddesses.* This is a late phenomenon, how-
ever, and by no means disproves the original identity of
the two.

In course of time the cult of the Phrygian Cybele pene-
trated North Africa, and was probably fused with the
cult of Tanith. The pressure also of the ascetic reaction
against the gross practices of this cult led, as it had done
at Petra,* to the representation of the goddess as a celes-
tial virgin.^ But we have Augustine's testimony that her
virginity was not of a very pure type. It is she, no doubt,
to whom Tertullian refers under the name of Ceres.'

The temple of Tanith-Dido was situated a little outside
the old city of Carthage, and in the fourth century of our
era was surrounded by a thorny jungle, which the popular
imagination pictured as filled with asps and dragons, the
guardians of her sanctuary.^ Outside its walls a pyre

^ It may be from the Semitic root in, "love," from which the name
David comes. Winckler thinks that in old Hebrew and Phoenician fn
was the genus loci, Greek Sal/iay. AUorient. Forschungen, pp. 339-
342.

2 ^nid, IV.

« Cults of the Greek States, Vol. II, pp. 638-642.

* Cf. Lidzbarski, Ephemeris fur semitische Epigraphik, Vol. I, p. 24.

5 Above, p. 233 ff.

« Augustine, De Civitate Dei, II, 4.

' Cf. Tertullian, Ad Uxorem, I, 6, and De Exhor. Cast., 15.

8 See the evidence cited in W. B. Smith's Seligion of the Semites, 2d
ed., p. 374, and especially Justin, XIX, 1.

256 SEMITIC ORIGINS

was erected each year, and the goddess was thought to
throw herself into the flames.^ As Virgil represents this
as done from love of vEneas, it was probably originally a
part of the mourning for her beloved Adonis, who was
evidently worshipped here.^ This topic will be treated
more fully below.

Passing now from the West to the East, we have some
of the oldest historical traces of the survival of this cult
at Erech, the ancient Uruk, modern Warka, in southern
Babylonia. Possibly the goddesses which survived at
Shirpurla are older, or equally old. These the exigencies
of our argument led us to treat in a previous chapter,^ so
that no more need be said of them here. The first dis-
tinct mention of Ishtar of Erech is in the inscriptions of
Lugalzaggisi, about 4000 B.C. If she is mentioned at all
earlier than that, it is by the kings of Shirpurla, and they
do not distinguish her from the Nana of their own city.
Lugalzaggisi calls her by the epithets Umu and Nina-
gidkhadu, calls her priestess and mistress of Erech.*
Later, if I do not misinterpret him, he calls Erech the
land of Ishtar.^ Something like twelve or fifteen hun-
dred years later Ur-Gur^ and Dungi,'^ of the kingdom
of Ur, repaired her temple. Ur-Ninib,^ Libit-Ishtar,^ and

1 See references in preceding note.

2 See the article " The Genesis of the God Eshmun," JAOS., Vol. XXP,
pp. 188-190. Above, p. 185 ff.

* OBI., No. 87, col. i, 11. 30-34; cf. Eadau, Early Babylonian History,
p. 183.

5 Ibid., col. ii, 1. 43 ff. It reads: KI NANA URUG-ki-i LU DAGAL
GUR-A-KIM MUR MU-DA-GIL, i.e. "The land of Ishtar, Erech, like a
sheep ready for shearing, I walled in with bricks." This reading presup-
poses that the determinative dingir has been omitted before Nana, and the
gunu signs from the ideogram for Erech. Such mistakes are not impos-
sible in this inscription: Cf. RU for NI, col. iii, 1. 37, and the variant.
Radau, op. cit., reads KI NANA URUG as Ki-inanni-ab.

6 I R., 1, No. 6. Cf. KB., Vol. IIP, p. 79.

' I R., 1, No. 3. Cf. KB., Vol. IIIi, p. 81. Also OBI., No. 15 ; cf. Radau,
op. cit., p. 226.

8 IV R., 35, No. 5. Cf. KB., Vol. IIIi, p. 85. Cf. also OBI, No. 18.
« I R., 3, No. 18. Cf. JTB., Vol. 111^ p. 87.

SURVIVALS 267

Ishmi-Dagan,! all of tlie dynasty of Isin, by repairing her
temple, or in some other way indicate their reverence for
her. About the year 2280 B.C. her temple at Erech was
destroyed by the Elamites, who captured a statue of her
and took it to Elam.^ It was probably the warfare con-
nected with this episode which has become the nucleus of
the Gilgamish epic.^

In the middle Babylonian period Karaindash (about
1450 B.C.) consecrated to her an inscription.* Later, in
the seventh century, Esarhaddon restored her temple and
worshipped her,^ and Assurbanipal brought back from
Elam the idol which had been taken thither from Erech
1635 years before.® In the next century Nebuchadnezzar
once more repaired her temple,'^ and after this time we lose
sight of her history. The temple upon which so many
kings had worked was called lanna, a name which it kept
from century to century.

These various sovereigns by their devotion attest the
importance of the goddess of Erech, but our chief source
of information concerning her is the Gilgamish epic.
This epic is composed of different strata which had their
origin in different periods and different centres, and which
are collected about the struggles of the hero Gilgamish,
against the Elamites, probably in the war of about 2280
B.c.^ In one of the oldest strata of the poem, Ishtar,
under the name Aruru,^ is represented as making a man
from a bit of clay as Yahwe does in the second chapter of

1 1 R., 2, No. 5, I and II. Cf. ^S., Vol. IIP, p. 87.

* V R., 6, 107 ff. Cf. KB., Vol. II, p. 209 ff.

» Cf. Haupt, Mmrodepos, pp. 24, 57, and KB., Vol. VI, p. 159 fi.

* IV R., 36, No. 3. Cf. KB., Vol. III^, p. 153.

s Cf. PA08., May, 1891, Sebraica, Vol. VIII, p. 113 &., and Vol. X,
p. 8ff.

^ See references under n. 2.

T Cf. I R., 65, col. II. 50 ff., KB., Vol. IIP, pp. 36, 37, and V R.,
34, col. II. 33, KB., Vol. nP, pp. 42, 43; also Hebraica, Vol. X,
pp. 12, 13.

' Cf. Jastrow, Beligion of Babylonia and Assyria, ch. xxiii.

' See Jastrow, Religion of Babylonia and Assyria, p. 448.
s

258 SEMITIC ORIGINS

Genesis.^ A little later this man, who is covered with
hair and is thoroughly wild, consorting only with beasts,
is enticed by intercourse with one of Ishtar's consecrated
harlots ^ to abandon his animals and to enter upon civilized
life. In this story, as already pointed out,^ we have the
survival of one of the primitive notions connected with
the Semitic mother goddess, viz. : the fact that civilization
arose from consciousness of sex. In appreciation of this
fact there were maintained for Ishtar, at Erech, three
classes of harlot priestesses.* Here, too, those rites which
Herodotus calls shameful ^ were also cherished.

In another part of this epic, which is also old, Ishtar
is represented as a deified woman of the early Semitic
times, who changed her husbands at will.® Tammuz, the
husband of her youth, she brought to mourning each year ;
various animals — for the myth originated in a totemistic
age^ — had been married by her, and through her had
come to grief. She then desired to wed Gilgamish, but
he, learning wisdom by what the others had suffered,
declined the honor. Such conceptions of her are another
proof of the character of the goddess which was fostered
at Erech. In order that Gilgamish should not escape her,
Ishtar had a bull created to torment him. These myths
which represent the goddess as such a harmful being,
probably embody in story the perception of the primitive
Semite that the unrestrained service of this sexual goddess
was fraught with physical peril.

1 Cf. Haupt, Mmrodepos, p. 8, KB., Vol. "VI, p. 121. Cf. Jastrow's
article "Adam and Eve in Babylonian Literature" in AJSL., Vol. XV,
pp. 193-214.

' Haupt, op. cit., p. 11; cf. KB., Vol. VI, p. 127. CI Hehraica, Vol.
X, p. .3.

8 Above, p. 84.

* Cf. KB., Vol. VI, pp. 176, 177 ; Jastrovr, Beligion of Babylonia and
Assyria, p. 475, and Haupt, Nimrodepos, p. 49, 11. 1, 2.

5 Bk. I, 199.

6 Haupt, op. cit., p. 42 ff. ; KB., Vol. VI, p. 171. Cf. Hehraica, Vol.
X, p. 5 ff.

' Above, p. 37.

SURVIVALS 259

The eleventh tablet of the epic contains the account of
the flood. This, as Jastrow has pointed out,^ originated
in connection with the city of Surippak, in quite a differ-
ent environment from the portions of the story we have
hitherto considered. As it now stands, however, it is
probably in mythology assimilated to the rest ; and in it
Ishtar is the mother of mankind who mourns that her off-
spring is destroyed.^

It is clear from a passage in the sixth tablet of the epic
already referred to,^ that the custom of wailing for Tam-
muz was a part of the ritual of the Ishtar cult at Erech.
It is possible that the myth embodied in the poem which
celebrates Ishtar's descent to the lower world originated,
or at least was cherished, at Erech. I formerly held on
mythological grounds that it was probably connected
especially with Nineveh,* but the same mythological data
would lead us to connect it with Khallabi^ and perhaps
with Shirpurla.^ As the Semitic settlements at these
latter towns are older than the settlement at Nineveh, it
can hardly have originated in the latter city. In the Gil-
gamish epic Ishtar is called daughter of Anu ^ instead of
daughter of Sin as in Ishtar's descent.^ The idea that she
was daughter of Anu may not have been the primitive
one, however, so that it is quite possible that the legend
of her descent to the lower world belongs to Erech in spite
of this. Jensen conjectured'* that the idea which finds
expression in the poem, that the underworld is surrounded
by seven walls, was, in the first instance, suggested by the
seven walls of Erech. Be this as it may, if the poem was
known at Erech, which is very probable, the conception
which it embodies that when Ishtar disappeared from the

1 Beligion of Babylonia and Assyria, p. 494 ff.

2 See Haupt, Nimrodepos, p. 139, 11. 117-125, and KB., Vol. VI, p. 239.

3 Above, p. 258. ^ See below.

* Hehraica, Vol. IX, p. 145, n. « See above, p. 199.

7 Cf. Haupt, op at., p. 46, 1. 107, and KB., Vol. VI, p. 173.

8 Cf. IV R., 31, 2 ; cf. KB., Vol. VI, p. 81.

* Kosmologie, p. 172 ff.

260 SEMITIC ORIGINS

earth, all desire in man and beast ceased,^ would be
another evidence of the survival at Erech in almost their
primitive purity of the earliest conceptions of Ishtar.

Another town where the cult of Ishtar flourished was
Khallabi, though but few references to her worship there
have as yet been recovered. From the fact that Khallabi
is mentioned on a contract tablet of the reign of Cambyses,
found at Abu-Habba, Jensen concludes that it was situated
near Sippar.^ We have but two glimpses of this worship, —
one in the reign of Arad-Sin, king of Larsa,^ and the other
in that of Khammurabi, king of Babylon.* Beyond the
fact that there was, in this city, a temple of the goddess
in which one of these kings placed a votive offering, and
which the other repaired, we are able to learn little of her
worship.

The ideogram for Khallabi occurs without the deter-
minative for place in one of the inscriptions of Gudea in
a passage descriptive of one of the goddesses of Shirpurla.*
The passage is a difficult one as both Jensen and Amiaud
recognize, but its meaning seems to be rendered clearer if
we understand the ideogram as Khallabi.^ The fact that
it lacks the determinative for place is not a serious objec-
tion to this interpretation, since Shirpurla itself is some-

iIV E., 31, 76 ff., and KB., Vol. VI, p. 87.

'^ Cf. Strassmaier's Cambyses, No. 48, 1. 2, and Jensen, KB., Vol. IIP,
p. 106, n. 6. Cf. also Zimmern, ZA., Vol. Ill, p. 97.
3 PSBA., Vol. XIII, pp. 158, 159.
* KB., Vol IIl\ p. 106 ff.

5 On statue r, De Sarzeo, Decouvertes, pi. 14, col. 1, 1. 16 ; cf. Amiaud,
Mecords of the Fast, New Ser., Vol. II, p. 98, and Jensen, KB., Vol. IIIi,
pp. 54, 55.

6 The passage would then read : (1. 12) i)jn</j>-GA-TUM-DU6 (13)
NIN-A-NI (14) SHm-PUR-LA-i:i (16) URU-KI-AG-GA-NI-TA (16) TE-
UNU-USLANUGUNU-ZA-A (col. ii, 1. 1) MU-NI-TU-DA-A (2)E-
di«£rir-GA-TUM-DUG (3) NIN-A-NA (4) RU-NI, i.e. "To Gatumdug,
his mistress for (the salie of) Shirpurla, the city which she loved (at)
Khallabi, where she bore him, the temple of Gatumdug, his mistress,
he built." On this view the ideogram for Khallabi has the sign TE in
addition to those which appear in it later, but for TE-UNU = UNU cf.
Brtinnow, List, No. 7721. For the order and identity of the other signs

SURVIVALS 261

times thus defectively written. ^ This passage makes it
probable that Khallabi was a colony or dependency of
Shirpurla in the earlier time, and that the worship of the
goddess reached this town through Shirpurla.

Passing northward from Babylonia to Assyria, the city
of Nineveh seems to have been another, town which re-
garded Ishtar as its tutelary divinity. Her connection
with this capital of the Assyrian empire is indicated by
the following facts : (1) The city was called in Assyrian
Ninua, a name which we have already seen to be a Semi-
tization through a folk etymology of the Sumerian Nind,
which was a part of the city of Shirpurla. ^ The identity
of the two names is shown by the fact that they are
expressed by the same ideogram. (2) By the fact that
Ishtar, the goddess of this capital city, is constantly asso-
ciated with Ashur, the god of the older capital from which
Assyria took its name, as one of the two leading deities
of the country. The chief temple in Nineveh, Ibarbar,
was a temple of Ishtar.^

The historical inscriptions of Assyria contain many
references to this goddess,* but beyond the fact that
she was considered to be " the firstborn of the gods "
(i.e. as the source and author of all life), and the
spouse of Ashur, these inscriptions contain little infor-
mation concerning her. The fact that a king like
Assurbanipal could import to Nineveh literature like
the poem of Ishtar 's descent, suggests that from his
youth he had known like traditions and like practices

of the ideogram, cf. Thureau Dangin's Mecherches sur Vorigine Vecriture
cuneiforme, p. 48. Gatumdug was, as we saw above, sometimes an epi-
thet of Bau, but here, of Nana.

1 Cf. De Sarzec, Decouvertes, pi. 31, No. 1, 1. 2 and pi. 2""-, No. 2, col.
1, 1. 3.

2 Cf. above, pp. 187 and 196 ff.

3 Cf. ZA., Vol. V, p. 79, 1. 40, and II R., 66, No. 2, 1. 9 ; Smith's
Assurbanipal, p. 305, 1. 9 ; also Ilebraica, Vol. IX, p. 143.

* These are collected in Eebraiaa, Vol. IX, pp. 132-143 and 156,
157.

262 SEMITIC ORIGINS

in connection with tlie worship of the goddess of his
native city.^

Arbela, a city of Assyria to the eastward of Nineveh,
was also the seat of an old shrine of Ishtar. The begin-
nings of her worship there are shrouded in a darkness
even more dense than that which covers the beginnings
of most of the Mesopotamian cities. We cannot even
guess from what part of southern Mesopotamia the immi-
grants who settled Arbela came. The honor in which
the Ishtar of Arbela was held toward the end of the
Assyrian empire seems to point to a considerable an-
tiquity for her worship there, but we obtain historical
glimpses of it only in the reigns of Sennacherib, Esar-
haddon, and Assurbanipal.^ By these kings the Ishtar of
Arbela is distinguished from the Ishtar of Nineveh. In
the earlier reigns, if such a distinction was made, it has
not been reflected in the literature hitherto recovered.

At Arbela, Ishtar continued to the close of Assyrian
history to be an unmarried mother goddess. As a mother
she was anxious for the welfare of her people, and conse-
quently ready to defend them against all their enemies.
Thus she became the goddess of war, to whom appeal was
naturally made in times of especial danger.^ In the
reigns of Esarhaddon and Assurbanipal she seems to have
been much sought after as the giver of oracles and the
revealer of the future fortunes of her worshippers.* Con-
nected with her temple at Arbela there seems to have
been an observatory from which astronomical reports
were sent to the king.^

These numerous and multiform survivals of the Ishtar

1 This view is confirmed by Maorobius (Saturnalia, I, 21), who speaks
of the worship of Adonis among the Assyrians as well as among the
Phoenicians.

2 Cf. Hebraica, Vol. IX, pp. 158-163.
^ Cf. Smith's Assurbanipal, p. 119 ff.

<Cf. IV R., 61, AJ8L., Vol. XIV, 267-277, and the references in
Hebraica, Vol. IX, pp. 160, 161.

^ See, for example, the texts in III R., 51.

SURVIVALS 263

cult, which project themselves far into civilizations which
could never have originated them, are, to him who has
an eye to read, sufficient evidence of the existence among
the primitive Semites of such a social order and such a
religion as that which is outlined in the second and third
chapters of this work.

It was suggested in a former chapter ^ that the primi-
tive Arabic environment and the early Semitic social
organization combined to create for the primitive Semitic
pantheon a goddess and her son. The transformations and
survivals of the goddess having now been traced, it remains
to inquire to what extent the primitive god, her son,
survived. In the course of the preceding argument it has
been frequently pointed out that, where the goddess
was transformed, the primitive god, her child, may have
been merged into the resultant god. In many portions of
the Semitic territory, however, his distinct survival can
be clearly traced.

Among the Nabathseans he survived as Dhu-'l-Shara.^
The polemics which, in this region as elsewhere, attended
the propagation of early Christianity, caused the fact to be
recorded that Dhu-'l-Shara was regarded as the son of the
old mother goddess.^ Of the cult of this god we have
little further evidence. The inscriptions which make
mention of him record the consecration of votive objects
or invoke his curse upon the violators of tombs. From
analogy we conclude that he was in character probabl}^
identical with Tammuz and Adonis, who were elsewhere
sons of kindred goddesses. This conjecture receives some
confirmation from the fact that Herodotus calls him
Dionysos.*

In Babylonia this god seems to have survived under
different forms. The Sumerian name by which he is most
commonly known is Dumu-zi, or Tammuz, " child of life."
Although this name is once applied to a goddess,^ it ordi-

1 Chapter III, p. 85 ff. ' Cf. Epiphanius, Panarion, LI.

2 CIS., Pt. II, Vol. I, No. 190-199. < Bk. Ill, 8. ^ Atove, p. 211 ff.

264 SEMITIC ORIGINS

narily refers to a god who is variously called the " offspring," ^
"the son of Ishtar," ^ and the "husband of her youth. "^
He was supposed to die periodically and was then bewailed.
Ishtar, once at least, and, perhaps, regularly, was thought
to go to the underworld for him and to bring him back to
life.* He is generally recognized as a god of vegetation ^
and of the underworld.®

At Shirpurla there was a god called Ningishzida, who
was associated with the goddess Bau,^ and who seems to
have possessed many of the characteristics of Tammuz.
In the Adapa legend Ningishzida and Tammuz are classed
together and play the same part ; both are intercessors for
the life of Adapa, and both are keepers of the gate of Anu.^
As pointed out above,^ Ningishzida seems to have been a
transformed Ishtar, but if so, Tammuz has been swallowed
up in him. The way in which, in Semitic antiquity, the
name of a deity constituted that deity a separate person-
ality is well illustrated by the way Tammuz and Ningish-
zida are in the Adapa story put side by side as two distinct
gods.

Jensen regards Tammuz as chthonic ; ^^ Jastrow regards
him as a solar deity. ^^ Probably both are right. If, as we
suppose,^^ he was associated by primitive Semites with
objects in an oasis, he was a chthonic god connected with
vegetation. The yearly death of vegetation and its sub-
sequent resurrection would, since it corresponded to the
movements of the sun, naturally lead in time to the identi-
fication of Tammuz with the sun. That would, however,
be a later view. In primitive times the god was chthonic.

In Phoenicia and the West the same god sur%dved under

I II R., 36, 54. 2 II R., 59, col. ii, 1. 9.

8 IV R., 31, 47, and Nimrodepos, p. 44 ; cf. KB., Vol. VI, pp. 168, 169.

* IV R., 31, and Jeremias's Leben nach dem Tode.

6 Jensen, Kosmologie, p. 197 ff. » Cf. KB., Vol. VI, pp. 96-99.

^ Jensen, op. cit., p. 225. ^ p. 190 ff.

' Above, p. 190 ff. i" Kosmologie, p. 197 fl.

^1 Religion of Babylonia and Assyria, p. 547 ff.
^2 Hehraica, Vol. X, p. 74, and above, p. 85 ff.

SURVIVALS 266

other names. Although Ezekiel called him Tammuz,i
Ezekiel lived in Babylonia and no doubt used the Baby-
lonian name or a corrupted form of it,^ — a name otherwise
unknown in the West.^ The clearest description of the
god which we have for this part of the Semitic world is
Lucian's account of the worship of Adonis at Biblos
(Gebal).* Here the myth had it that the god was killed
by the tusk of a boar, and the reddening of the river by
the highly colored soil was held to be the result of shed-
ding the blood of the god. The sexual rites connected
with his worship at this place make it very clear that his
cult was a survival from primitive Semitic times. Later,
the myth was interpreted as a nature myth, the tusk of
the boar being regarded as the inclement winter and the
resurrection of the god as his victory over the first six
signs of the zodiac.^

The name Adonis, by which Lucian designates this deity,
is simply the Semitic epithet Adon, "Lord"; it was not
the real name of the god. As I have pointed out else-
where,^ the real name of this deity was probably Eshmun.
The reasons for considering Eshmun as a Tammuz are :
1. That the epithet Adon was frequently applied to Esh-
mun, as the name Eshmun-Adon, which was quite popular,
shows. 2. Eshmun was a very popular god among the
Phoenicians, — as popular as one would expect Tammuz to
be. 3. Eshmun was a god of the healing art and was
identified with the Greek ^sculapius.'^ Several scholars
identify him also with lolaos, who in a Semitic myth in
Greek dress, saved the life of Heracles.^ Similar charac-

1 Ch. 8". " Dumuzu or Duzu. Cf. atove, p. 263.

' Cf. Baethgen, Beitrdge zur semitischen Heligionsgeschichte, p. 44.
Cf. also JAOS., Vol. XXP, p. 188.

* De Syria Dea, §§6, 8 ; quoted above, p. 245.

» Macrobius, Saturnalia, I, 21. « JAOS., Vol. XXP, p. 188 ff.

' CIS., Pt. I, Vol. I, No. 143. On ^sculapius, cf. Dyer's Gods of
Greece, pp. 220-256.

8 See W. R. Smith, Beligion of the Semites, 2d ed., p. 469, and Pietsch-
Biann, Phcenizier, p. 181.

266 SEMITIC ORIGINS

teristics seem to have pertained to Tammuz. If Jeremias
is correct in his interpretation of the enigmatical lines at
the end of the poem on Ishtar's descent to the lower
world,^ appeal could be made, on the day of Tammuz, for
the restoration of the dead to life. Such restoration was
but a heightened form of healing the sick. 4. As the
worship of Tammuz was closely connected with the wor-
ship of Ishtar-Ashtart, so was the worship of Eshmun ;
and also as we should naturally expect with that of Baal,
the transformed Ashtart. At Carthage, Tanith-Ashtart
and Baal were worshipped in the temple of Eshmun;^
while Hannibal in ratifying the treaty with Philip of
Macedon, swore by Heracles (Baal) and lolaos (Esh-
mun) .^ Once Eshmun and Ashtart are compounded into a
single deity,* — a fact which points strongly to a conscious-
ness of identity of function, — an identity which in turn
points to kinship of origin. At Si don his worship was
very popular and took rank with that of Baal and Ashtart.^
At Kition and Idalion in Cyprus, where there were im-
portant temples of Ashtart,^ the worship of Eshmun also
flourished. This is proven by the many proper names on
the monuments from these places, into which the name
Eshmun enters as an element. Eshmun was, in the same
region, called "Melqart," as several inscriptions show.^
This title, meaning " king of the city," applies usually to
the Tyrian Baal, and its application to Eshmun probably
indicates a conscious unioia of Eshmun with that Baal.^
Such a union with Baal, like the union with Ashtart, points
to a similarity of function, and consequently of origin, for
the two gods. As Baal was a transformed Ashtart, this

1 Cf. Leben nach dem Tode, p. 7. ^ Polybius, vii, 9, 2.

2 cm, Pt. I, Vol. I, No. 252. ' CIS., Pt. I, Vol. I, No. 245.
^ Ibid., No. 3.

6 Cf. CIS. , as above, No. 86, Journal of Hellenic Studies, 1888, pp.
175-206 ; Tacitus, Historia, II, 2, 3 ; Pausanias, viii, 6, 2, etc.

' CIS., Pt. I, Vol. I, Nos. 24-28.

8 See the article "West Semitic Deities with Compound Names" in
JBL.,Vo\. XX, p. 22 ff.

SURVIVALS 267

fact points ultimately to a primitive Semitic origin for
both Ashtart and Eshmun, — or in other words to the fact
that Eshmun was a Tammuz. 5. With Eshmun as JEscu-
lapius there are associated two accounts^ of exposure to
death and deliverance from it which approximate the
death and resurrection of Tammuz. These accounts are
probably variant versions of the myth which Lucian tells
of Adonis at Gebal. In all probability, therefore, the
god who is called Tammuz in Babylonia, Dhu-'l-Shara in
north Arabia, and was known to the Greeks at Gebal as
Adonis, was known among the Phcsnicians generally as
Eshmun.^ This variety of name is due to the fact that
the primitive Semitic appellation did not survive.

The myth of the death of the deity sometimes attached
to the primitive goddess rather than to the god. This was
apparently the case at Carthage, where Dido, it was said,
yearly threw herself to death. ^ Such a variation is another
indication of the close connection between the primitive
goddess and her son.

In primitive Arabia the conditions of the country would
produce two kinds of clans. On the oases the communal
clan, devoted to the worship of the mother goddess, would
flourish.* The flocks and the caravan trade would lead to
the organization of the republican clan, the more hazard-
ous life of which would lead to the worship of the ideal
masculine character. Such clans are found in Arabia
to-day,^ and doubtless were called into existence in prehis-
toric times by the peculiar economic character of Arabia.

1 Cf. Pausanias, II, 26*^.

2 What the name of Eshmun means and how it originated, it is hard
to say. It was probahly originally some kind of an epithet. Of the sug-
gestions made, the one most worthy of credence is, perhaps, that of
Lagarde {Qr. Uebers. der Prow., p. 81), and repeated by W. R. Smith
(lieligion of the Semites, 2d ed., p. 469), viz.: that the name is to be

connected with the Arabic jL^-w/, "quail," because in the myth lolaos
brought Heracles to life by giving him a quail to smell of.

8 Cf. "W. R. Smith, lieligion of the Semites, 2d ed., pp. 374, 410, and
above, p. 256 fl. * See above, p. 30. * Cf. above, p. 72, n. 1.

268

SEMITIC ORIGINS

Among such clans the worship of Tammuz would be prac-
tised, and as they migrated it would be diffused. Depend-
ent as these clans were upon the oases for much of their
nourishment, they would long perpetuate the myth that
their god was the son of the goddess of the oases. The
survival of Tammuz, Dhu-'l-Shara, Adon, and Eshmun
was most natural. The widespread cult is accounted for
by primitive Semitic social conditions, and in turn is an
important link in the chain of evidence which enables us
to restore the outlines of that far-off social and religiouA
organization.

CHEONOLOGICAL TABLE

El-Amama tablets, oir. 1400 b.c.
Exodus from Egypt, cir. 1250.
Israel Invaded Canaan, cir. 1200.
David became king, cir. 1000.
Kingdom divided, 937.
Ahab king of Israel, 876-854.
The Prophet Amos, cir. 760.
The Prophet Hosea, oir. 745.
The Prophet Isaiah, 740-700.
Fall of Northern Kingdom, 722.
Manasseh king of Judah, 696-641.
Josiah king of Judah, 639-608.
Jeremiah the Prophet, 627-cir. 580.
Josiah's reform, 621.
The Prophet Ezekiel, 593-570.
Destruction of Jerusalem, 586.

The second Isaiah, cir. 545.

Cyrus captured Babylon and per-
mitted return of Jews, 538.

Second Temple completed, 516.

Nehemiah, governor of Jerusalem,
444.

Jews pass under Greek rule, 332.

Jews under Egyptian Ptolemies,
323-198.

Jews pass finally under Seleucids of
Antiooh, 198.

Earliest parts of Enoch, 200-170.

Maccabsean revolt, 168-165.

Jews independent under Simon and
the Asmonseans, 143-63.

Judea passes under Roman sway, 63.

CHAPTER VII

TAHWE

In sketching the transformations which the primitive
Semitic goddess underwent^ as the Semites wandered
from their Arabian home into other environments, no
mention was made of Yah we, the God of Israel. This
omission was not accidental. No deity of the old Semitic
world compares in importance with Yahwe ; the worship-
pers of no other god contributed to the sum of humanity's
ethical ideas and spiritual conceptions a tithe of the value
of that contributed by the worshippers of Yahwe. The
importance of Yahwe, therefore, demands that a separate
chapter should be devoted to him. It is evident to one
who has followed with any sympathy the argument of the
preceding pages that the religion of the Semites as a whole
moved forward by a process of evolution in which it was
subject to certain great principles of general application.
Is the religion of Israel subject in any degree to these
great principles ? Is its God Yahwe connected at all with
that primitive Semitic root from which we have found
nearly all other Semitic gods to spring ? If he is, can it
be claimed that there is in the Old Testament any special
revelation of permanent religious value ? These are ques-
tions which we must now try to answer, and it is the
writer's belief that to each one of these an affirmative
reply can be given.

The critical study of the Old Testament, which has
seemed to some to destroy the historical and religious
value of the earlier books of the Bible altogether,^ has

1 Above, Chapters IV and V.

2 The reader who may chance to be unfamiliar with the results of
criticism will find compendious statements of it in Driver's Introduction

269

270 SEMITIC ORIGINS

really opened a new historical vista to the student of any
phase of Israelitish history. This is as true of Israel's
religion as of any other phase of her life. While there
are critics who can bring themselves to regard as histori-
cal scarcely any of the material which relates to the times
before David and Solomon, most critics regard the broad
outline of the traditions which relate to the sojourn in
Egypt, the exodus, the wilderness sojourn, and the con-
quest of Canaan, as representing real facts of history.
This does not imply that there is no need to apply critical
methods to these traditions in order to ascertain the truth.
Tradition has no doubt often destroyed the historical per-
spective ; it has applied to the whole of the nation that
which in reality belonged only to parts of it. A discrimi-
nating student can nevertheless still in good degree un-
tangle the thread and restore the main features of the
history. In this task much help is secured by the recog-
nition of the simple fact that in the genealogies tribes are
often personified as men.

The beginnings of the nation Israel may, by the aid of
critical study, be broadly sketched as follows : ^ From
time immemorial wave upon wave of Semites had over-
run Palestine, and had by fusion with its aboriginal
inhabitants, whatever they were,^ gradually formed the

to the Literature of the Old Testament, or Cornill's Einleitung in das
alte Testament. The best presentation of the criticism of the Pentateuch
and Joshua is Carpenter and Harford-Battersby's Hexateuch.

1 This sketch gives the outline which my own studies have led me to
adopt as most probable. For sketches of other scholars, cf. Kuenen,
Religion of Israel, pp. 109-115 ; Stade, Geschichte des Volkes Israel,
Vol. I, p. 113 ff. ; Wellhausen, Israelitische und judische Geschichte,
oh. ii ; Kent, History of the Hebrew People, Vol. I, ch. v ; Cornill,
History of the People of Israel, pp. 45-55; Guthe, Geschichte des VolJces
Israel, pp. 12-28, and § 9 of his article "Israel," in Encyc. Bih.;
Winckler, Geschichte Israels, Vol. I, pp. 12-24 ; and § 1 of Woods's
article, " Israel," in Hastings's Dictionary of the Bible.

2 Sergi (Mediterranean Race, pp. 150-156) believes that these aborigi-
nes were Hittites who had separated from the North African race. It
must, if true, have been at a considerably earlier time, of course, than
the Semitic migration from Africa.

YAHWE 271

Phcenician or Canaanitish peoples. It has been already
pointed out^ that from the time of Lugalzaggisi (about
4000 B.C.) onward, many successive expeditions of con-
quest and migration from Babylonia had also swept over
the land. With these Babylonians there were mingled,
from about 1500 B.C. onward (and for all we know, from
a much earlier period), Aramsean tribes who had pre-
viously inhabited the highlands between the Mesopotamian
valley and the Mediterranean. The presence of these
tribes can be traced in the El-Amarna letters about
1400 B.c.^ A number of the clans which were afterward
united into the nation Israel belonged to this Aramaic
group of nomads. This is proven by the persistent tradi-
tions which connected Hebrew ancestry with Aram,^ and
receives confirmation from phenomena in the El-Amarna
letters, which will soon be noted.

Of these clans, the Reubenites may have been at first
the chief,* but that leadership soon gave way to the power-
ful Joseph clan, later divided into the clans of Ephraim
and Manasseh, and of which the clan of Benjamin was a
later offshoot. ^ Closely allied to the Reubenites were the
clans of Issachar and Zebulon, and less closely the clans
of Gad and Asher, the last of which we have already
traced in Palestine in the El-Amarna period.®

1 See above, pp. 146 and 160.

' They are mentioned in the annals of Tiglath-pileser I, about 1110 B.C.
Cf. KB., Vol. I, p. 33, and "Aram," in Mncyc. Bib. and Jevyish Encyclo-
pedia, and above, p. 226 ff.

3 Cf. Gen. 12*, 24, 281-322, and Dent. 266.

* Cf. the tradition that he was Jacob's firstborn, Gen. 29'^, 49', etc.
Birthright implied hegemony and power.

6 Cf. the tradition of Benjamin's late birth, Gen. SS^"*-!*. The name
Benjamin is really Bne-Yamin, "sons of the south," i.e "southerners."
The kinship to the Joseph tribes which the traditions assign means that
the Benjaminites were the "southerners" of the Josephites. Cf. the
Arabic Yemenites. A more remote kinship to these tribes is assigned by
the traditions to the tribes of Dan and Naphtali. This means that they
joined the confederacy later, perhaps after the emancipation from Egypt.

^ The tribes of Judah, Simeon, and Levi are also assigned in the tradi-
tions to the Leah group. On Judah, see p. 272, n. 4. Simeon as a clan

272 SEMITIC ORIGINS

The Joseph clans ^ wandered in time of famine to
Egypt,^ whither they were followed by others, probably
some of the Leah tribes, of which the Reubenites were the
most powerful. There, in course of time, these tribes
found themselves in bondage. Meantime the Kenites, a
clan whose origin was more directly Arabian, having been
touched by the northern wave of Mineean influence from
south Arabia,^ and which afterward formed part of the
tribe of Judah,* had occupied the Sinaitic peninsula and
the region to the north of it, and had become a pastoral
people. Moses, a man of one of the tribes which were in
bondage, fled from Egypt, sojourned among the Kenites,
became a devotee of the Kenite god, Yahwe, went back to
Egypt, proclaimed Yahwe the deliverer of the oppressed
clans, led his brethren to Sinai, and with the aid of Jethro,
Yahwe's priest in Midian, bound them for the future in
alliance with the Kenites and to the service of Yahwe. ^

has in tlie historical period a most shadowy and problematical existence.
Possibly it was an early clan which was overtaken by misfortune (Gen.
495 ff). The same may be said of the tribe of Levi, though it is possible,
as Budde thinks {Bel. of Israel to the Exile, p. 80 ff.), that it was a
priestly clan of later origin.

1 Cf. Wildeboer, Jahvedienst en Volksreligie in Israel, p. 15.

2 Winckler, Altorientalische Forschungen, p. 337 ff. , has suggested
on the basis of a Sab^an inscription that Misraim is in the Old Testa-
ment a later misunderstanding for Misr, a name which he believes was
applied to the part of Arabia which included the Sinaitic peninsula, and
that the Hebrews never were in Egypt proper. This view is accepted by
Schmidt, American Journal of Theology, Vol. V, p. 136. This is, how-
ever, too slight a basis on which to cast away all the traditions of later time.

' Cf. Weber's article, " Studien zur siidarabisohen Altertumskunde,"
and the literature cited in it, published in Mitthelungen der vorderasiat.
Gesell., 1901, especially pp. 29 and 36 ff. Cf. also Lidzbarski's Ephem-
eris fur semitische Epigraphilt, Vol. I, p. 128.

* The traditional genealogies indicate that there was a close kinship
between Judah and the Reubenites. This means that there was an
Aramaic element in Judah, to which other elements, as pointed out below,
were joined later. The possible mention of Judah in the El-Amarna
tablets (cf. Jastrow, JBL., Vol. XII, p. 61 fi.) would, if real, seem to
confirm this view.

6 Cf. Ex. 1812 ff- and Budde, liel. of Israel to the Exile, p. 22 ft.

YAHWE 273

After wandering for a time as nomads, these clans or a
part of them conquered the east-Jordanic country, in
which probably some of their kinsmen, the tribe of Gad,
had remained from earlier times. After they became too
numerous for this region, the Jordan was crossed and the
heart of Palestine conquered. After their settlement
there, tribes which had never left the country, such as
Asher and perhaps Dan and Naphtali, were incorporated
with them. It is perhaps true that an Aramaic element,
kindred to the Reubenites, an element which formed the
nucleus of the tribe of Judah, was concerned in this general
movement to Egypt and back ; but the Kenite clan, at
least in part, and perhaps others who were afterward im-
portant elements of the tribe of Judah, moved from the
south into the territory later occupied by them making
their entrance at a time considerably subsequent to that
of the Joseph tribes. Long after the time when the book
of Judges takes up Israel's history, Judah was even less
closely attached to the other tribes than they were to one
another.

These Israelitish clans — always in the early days with-
out fixed organization — became in time, by absorbing
elements alreadj'' in the land, the tribes of the book of
Judges. Within each tribe there seems to have been no
more organization than Arabic tribes in the desert show
at the present time, and as regards one another they had no
real governmental connection. A sense of kinship and of
loose alliance was their only bond. The two most power-
ful of these were the clans of Joseph and Judah. These
clans were never permanently united, and afterward
formed the centres of the northern and southern king-
doms. Some of the features of this sketch will be en-
larged upon below when some of the proof for it will be
considered. At present we must turn to one or two
matters which are thought to oppose difficulties to the
course of events just outlined.

Among the El-Amarna letters, about 1400 B.C., there

274 SEMITIC ORIGINS

are several from Abdikheba of Jerusalem ^ in which he
complains that his government is being overthrown by a
people called Khabiri, whom Zimmern and Winckler^ have
identified with the Hebrews. If this identification were
correct, it would follow that the exodus should be dated
considerably earlier than has of late been customary among
scholars. That the Khabiri and the Hebrews are the
same is, however, very improbable.^ The suggestion of
Jastrow* that the Khabiri were a clan afterward embodied
in the tribe of Asher as Heber (Kheber) seems to me far
more probable. If, as we have supposed,^ the tribe of
Asher was fused with the other tribes after the settlement
in Canaan, the presence of the Khabiri about Jerusalem at
a time when the bulk of the Hebrews were in Egypt would
afford no difficulty.

Another fact which is supposed by some to present diffi-
culty is the mention of Israel on a stele of Meren-Ptah
(Menephtah), discovered by Petrie in 1896. The context
places Israel among enemies whom the king destroyed in
Palestine.^ This implies that Israel was settled there in
the reign of the Pharaoh under whom the exodus is
usually supposed to have occurred. The force of this
consideration some would break by the claim that the
poetical and exaggerated language of the inscription of
Menephtah cannot be sufficiently definite to be taken
seriously. It seems clear, however, that if the inscription
has any meaning at all, it implies that Israel was in Pales-
tine when it was written. But it does not follow that the
term " Israel " then connoted all that it signified at a later
time. Jacob and Joseph in the reign of Thothmes III,'^ a

1 Cf. KB., Vol. V, Nos. 179-185.

^ See his Geschichte Israels, Vol. I, pp. 17-20.

8 Cf. Hommel's Ancient Hebrew Tradition, pp. 230 ff. and 258 ff.

4 In JSL., Vol. XI, p. 120. ^p. 273.

6Cf. Steindorf in ZATW., Vol. XVI, p. 330 ff.; and Breasted in
Biblical World, Vol. IX, p. 62 ff. For a summary of conflicting opinions
cf. ibid.. Vol. VIII, p. 243 ft.

' W. Max Miiller, Asien und Europa, p. 168.

TAHWE 275

little earlier, were the names of places, and Joseph at
least underwent a change ; may not Israel have done the
same ? The inscription makes it clear that Israel is used
in the sense of a people ; but if our view of the gradual
aggregation of the Israel of later times be correct, not all
the clans which we know under that designation in the
Old Testament need have been present among those whom
Menephtah vanquished. If a small detachment were there,
the conditions would be satisfied. ^ This difficulty there-
fore vanishes.^

On the view of the origin of the Israelites outlined
above, Yahwe was the god of the Kenites before he became
the God of Israel. This view was first suggested by
Ghillany,^ and afterward independently by Tiele,* more
fully urged by Stade,^ and has been thoroughly worked
out by Budde.® It is now accepted by others, as Guthe,''
Wildeboer,^ and H. P. Smith. ^ It is naturally rejected
by Dillmann,^'' and his school,^^ as well as by writers like
Robertson,i2 -^i^q contend against all critical theories of
the origin of Israel's religion. The reasons for accepting
the view that Yahwe was the god of the Kenites before
Moses mediated a covenant whereby he became the god of

1 Cf. Budde, Bel. of Israel, p. 7.

2 The indefiniteness of Menephtah's use of the word " Israel " is shown
by the contradictory theories built upon it ; cf. Biblical World, Vol.
VIII, p. 243 ff.

8 In 1862, writing under the pseudonym of Richard von der Aim.
Cf. Holzinger in Exodus in Marti's Kurzer Hand-Gom., p. 13.

* Manuel de Vhistoire des religions, 1880, p. 84 ; Histoire comparee
des anciennes religions, 1882, ch. ix; and Outline of the History of
Ancient Eeligions, 1888, p. 85.

s Oeschichte des Volkes Israel, Vol. I, p. 130 ff.
« New World, 1895, pp. 726-746 ; and Meligion of Israel to the Exile.
' GesrMchte des Volkes Israel, p. 21.
^ Jahvedienst en Volksreligie in Israel, p. 15 ff.
' American Journal of Theology, Vol. IV, p. 549 ff.
w Com. on Exodus (ch. 3") ; and Alttestamentliche Theologie, p. 103, n.
" Cf. e.g. Kittel, History of Israel, Vol. I, p. 250 ; and Strack, Com. on
Exodus (ch. 3").

12 See his Early Religion of Israel.

276 SEMITIC ORIGINS

the Joseph tribes and ultimately of Israel are, in brief, as
follows : —

1. Of the three documents, J, E, and P, which narrate
the exodus, two, E and P, relate that the name Yahwe
was quite unknown until the time of Moses,i and that it
was revealed to him while tending the flock at Yahwe's
mount of Horeb or Sinai. Moses was told that he was
treading on holy ground, i.e. that the mountain where he
was was the sacred dwelling of Yahwe. P declares that
the patriarchs had worshipped Yahwe under the name
El-Shaddai, but that he was unknown to them by his
name Yahwe. It was thus by the late date at which P
wrote that the identity of two gods could be asserted, but
in the earlier time of Moses such was not the case. A
different name soon came to mean a different deity, even
when it had been at first a mere epithet of a god already
well known. E, on the other hand, declares^ that up to
the time of the exodns the ancestors of Israel had been
idolaters. True, he seems to make an exception in the
case of Abraham, Isaac, and Jacob,^ but the exception is
more in seeming than in reality. E, as critics agree,* was
an Ephraimite. In him the traditions as they were current
among the Joseph tribes find expression ; and those tradi-
tions had preserved the definite recollection that the
knowledge of Yahwe was not original in Israel, but came
in at the time of Moses. P was dependent for his knowl-
edge of the subject upon E, and simply retold the story in
his own way.

2. That Yahwe was the god of the Kenites is further
shown by the nature of the sacrificial covenant which,
according to E,^ preceded the giving of the law. At that
sacrifice to Yahwe it was not Moses or Aaron who offi-
ciated as though initiating Jethro into a new worship, but

1 E, in Ex. 313 ff-; P, in Ex. 6'«: 2 Josh. 24". «Ex. SK

<Cf. Kuenen, Mexateuch, p. 248 ft.; Driver, Introduction, p. 116 ff.;
and Carpenter and Harford-Battersby's Hexateuch, Vol. I, p. 116.
'Ex. 18i2ff-. Cf. Budde, Religion of Israel, p. 22.

YAHWE 277

Jethro, the Kenite,^ officiated as though introducing Moses
into a new cult.

3. For centuries after Moses Sinai was regarded as the
home of Yahwe, even when it lay beyond Israel's borders.
From Sinai Yahwe came to give victory to his people in
the days of Deborah ; ^ to Sinai Elijah made a pilgrimage
in order to seek Yahwe in his home ; ^ and the prophetic
writer who shaped the blessing of Moses echoed the same
conception.* So deeply was the idea fixed in the religious
thought that it survived in poetry in post-exilic times after
the sanctity of the temple at Jerusalem had caused that
structure to supplant Sinai in the popular thought as the
abode of Yahwe. ^

4. The Kenites were during several succeeding centu-
ries the champions of the pure worship of Yahwe, even
among the Hebrews themselves. Thus Jael, the wife of
Heber, the Kenite, was the slaughterer of Sisera and the
champion of Yahwe ; ^ Jonadab, the son of Rechab, who
was a Kenite,' and who maintained the nomadic ideals of
the worship of Yahwe as they had existed in the steD|M^
aided Jehu in the eradication of Baal-worship in Israel, ^d
in the establishment of the worship of Yahwe ; ^ and centu-
ries later the fidelity of these Kenitic Rechabites to Yahwe
was such that it served admirably in the hands of Jere-
miah to point a moral to his degenerate fellow-citizens.®

5. These Kenites (sometimes called Midianites ^'') seem,
a part of them, to have joined Israel in their migrations,^!
becoming mingled with the people at various points, both
in the North ^^ and the South,!^ and in part they remained in
their old habitat on the southern borders of Judah as a
separate though friendly clan in the days of Saul ; i* finally,

1 Cf. Jnd. 115 and 4". * Deut. 332. 7 cf. 1 Chr. 2^^.

' Jud. 5* «■-. 6 Cf . Hab. 31 and Ps. 68^ Cfl. s 2 Kgs. W^.

« 1 Kgs. 19. « Jud. S^iff- and 4i'ff-. » Jer. 35.

^ The Kenites seem to have been a part of the Midianites. The latter
was the broader term. Cf. Budde, Bel. of Israel, p. 19.

11 Num. 10" ff-. 12 Jud. 4"ff; 524ff-. is Jud. l". " 1 Sam. 156.

278 SEMITIC ORIGINS

in the days of David, they were incorporated into the tribe
of Judah,^ with which they were afterward counted. ^

6. Now it was in the tribe of Judah, into which these
Kenites had been incorporated, that, as most recent critics
believe,^ the J document was composed, — that document
which betrays no consciousness tliat there had ever been a
time in Israel when the worship of Yah we was unknown,
and which makes that worship almost coeval with man.*
This fact is all the more striking when it is remembered
that J had accepted so many of the Aramaic traditions^
which were in all probability originally the possessions of
the tribes farther to the north ; and it is best accounted
for by the supposition that the Kenites, whose god Yahwe
originally was, had been fused with the tribe of Judah
and had thus infused into Judsean tradition a strong semi-
Arabian current of thought, on which was borne the con-
sciousness of the immemorial knowledge of Yahwe. The
perpetual separateness of Judah from the other tribes
would help to maintain this tradition in spite of antago-
nistic currents from other quarters.

We conclude, therefore, that the result of the application
of critical methods to the history of Israel is to make it
clear that Yahwe was the god of the Kenites before the
days of Moses.

Can we now go farther and determine anything of the
nature of Yahwe or of his history before he became the God
of Israel ? Our investigation has, I think, placed us in a
position to do this. But before proceeding to the task
we must first notice a view which has sometimes been ad-

11 Sam. 3026ff>"P.29.

2 Cf. the genealogy of the Calebites and Bethlehemites in 1 Chr. 2,
ending with v. 55, according to which David himself came from a family
of Kenites. For a comprehensive statement of the Biblical data concern-
ing the Kenites see Kuenen's Beligion of Israel, pp. 179-182. The state-
ment is condensed from an earlier article of Nbldeke.

* See discussions in Driver's Introduction, p. 115 ft., and Carpenter and
Harford-Battersby's Hexateuch, Vol. I, pp. 104-106.

* Gen 426. 6 cf. Gen. 24 and the J element in Gen. 28-31.

YAHWE 279

vocated,^ and which I formerly held,^ that Yahwe in his
primitive character was a storm god. In favor of this the-
ory it may be urged that in the theophanies he is usually
represented as coming in a thunder-storm ; ^ that he is said
to have led his people in a cloud ; * that he appeared on
Mount Sinai and in the temple as a cloud ; ^ that in the
middle books of the Pentateuch the cloud is used as a
token of Yahwe's presence more than forty times ; that
the thunder was the " voice of Yahwe " ; ^ and that Yahwe
controlled the stormy movements of nature.^ These facts,
which are beyond dispute, have led Winckler to regard
Yahwe as a Hadad or a Ramman.^

There can be no doubt but that in the case of Yahwe, as
in that of Hadad and Ramman, the god was conceived as
controlling the phenomena of the weather and of the
heavens, and of manifesting himself through them. Such
conceptions may well have been entertained by residents
in the Sinaitic region as well as by the Aramaeans, resi-
dent in the various parts of Syria, and by the ancient As-
syrians and Babylonians. Robertson Smith has, however,
wisely warned us against finding the origin of any Semitic
god in the personification of any one power of nature ; ^
the primitive Semite looked to his god to perform for him
the whole circle of divine activities, and the theory that
Yahwe was primarily the personification of the storm is as
inadequate as the theory that Hadad or Ramman was.^"
Indeed, we are now in a position to show that in all prob-
ability the Yahwe of the Kenites was developed like

1 Cf. Stade, aeschichte des Volkes Israel, Vol. I, p. 429 ff.

2 Cf. Oriental Studies of the Oriental Club of Philadelphia, p. 86 fl.
« Cf. Ps. 18, Ez. 1, Hab. 3, Isa. 19i, and Job 38i.

4 Ex. 13 and U.
= Ex. 19 and 1 Kgs. 8i»".
« Ps. 293 «, Job 37^ and Ps. 68".
' Ps. 10418- 14^ and Ps. 1478- i«-i8.
' Geschichte Israels, Vol. I, p. 37 ff.

^Religion of the Semites, 2d ed., p. 81 fl. Cf. Budde, Bel. of Israel,
p. 57, n.

i» See above, p. 227 ff.

280 SEMITIC ORIGINS

Ramman, Hadad, and most other Semitic deities, by the
same processes which we have traced elsewhere, out of the
primitive mother goddess. The reasons for this view are
as follows : —

1. The Kenites were a Semitic tribe resident upon the
confines of Arabia itself, and it is to be presumed that
their religious life had been continuous and subject to the
ordinary laws of Semitic development. 2. They were a
pastoral people,^ who, according to the general laws of clan
organization as outlined by Professor Keasbey,^ must have
developed the patriarchal clan. Like the Moabites,^ their
neighbors, then, the primitive goddess which was their
common Semitic inheritance had among them been trans-
formed into a corresponding masculine deity. 3. That
Yahwe had some genetic connection with the primitive
goddess is shown by the emphasis which his cult laid upon
circumcision. We are told by J* that Yahwe sought to
kill Moses till his son was circumcised, when the god be-
came friendly.* The same writer tells us that after their
entrance into Canaan ^ the marriageable young men were
circumcised to complete their consecration to Yahwe.
The Priestly writer represents circumcision as instituted
in the time of Abraham as a token of Yahwe's covenant
with him,^ and informs us that no uncircumcised person
could keep Yahwe's passover.^ Such was the stress laid
upon circumcision that in later times it became a synonym
for Israelite, and uncircumcised a synonym for foreigner.*
Circumcision became also a synonym for all the spiritua^
and ethical qualities for which the Yahwe cult had then

1 Ex. 2W «••. 31 «: 2 Above, p. 30. « Above, p. 140 ff. * Ex. 4 ^*- ^.

5 The real meaning of the passage seems to be that Moses himself was
uncircumcised, and that, therefore, Yahwe tried to kill him ; that Moses's
wife circumcised her son and smeared the blood upon Moses, so as to
make it appear that the blood proceeded from an incision in him, and
that then Yahwe was appeased. Cf. Wellhausen, Heidentum, 2d ed.,
p. 176.

6 Josh. 53- 9. ' Gen. 17. ' Ex. 12*8.
9 Cf. 1 Sam. 31«, 2 Sam. V>, 1 Chr. 10*, and Eom. S^.

YAHWE 281

come to stand. 1 Abraham, it was thought, would save
from the pit all who bore the mark of circumcision.^ How
deeply fixed this rite became is indicated by the struggle
which Paul and others had to undergo in order to throw
it off. The fixed and important character which it had at
all periods indicates that from the very beginning it must
have been considered a vital part of the religion of Yahwe,
and must have had its motive in a conception which iden-
tified the rite with some of Yahwe's most important func-
tions. Now, at the first, circumcision seems to have been
in Israel itself a preparation for connubium.^ The same
rite with the same meaning we have previously found to
be a part of the cult of the primitive mother goddess.*
The existence of circumcision in the cult of Yahwe is
therefore a strong argument for the theory that the cult
of Yahwe was a direct development from that primitive
Semitic worship.

4. Another indication that Yahwe was originally devel-
oped out of the mother goddess is the old Hebrew custom
of swearing by Yahwe with the hand " under the thigh,"
i.e. upon the organs of reproduction.^ This custom shows
that in early times this part of the body must have been
especially sacred to Yahwe. That would naturally be the
case if he were developed from an Ashtart. 5. All critics
agree that the passover was the feast of Yahwe which
without question antedates the settlement in Canaan.^
This festival, with its sacrifice of a sheep, we have already
traced^ in its beginnings to the feast of the primitive

iCf. Kom. 228ff-.

» Cf . Weber's Judische Theologie, 2d ed., pp. 342, 343.
*Cf. Gen. 34 and Ex. 4^5. In the latter passage the phrase "bride-
groom of blood" connects it with connubium.

4 See above, pp. 98 ff. and 110 ff.
6Cf. Gen. 242-9, and 47^9.

5 Cf. e.g. Wellhausen, Prolegomena, 5th ed., ch. iii ; W. R. Smith, Bel.
of Sem., 2d ed., 227 ft., 445 ff.; Piepenbring, Theology of the Old Testa-
ment, p. 50 ; Budde, Bel. of Israel, p. 73 ff., and Moulton's article "Pass-
over " in Hastings's Diet, of the Bible.

' Above, p. 109 ff.

282 SEMITIC OUIGINS

Semitic goddess. If, therefore, it was a festival of Yahwe
in the steppe, it is another link connecting him with that
primitive cult. This inference receives confirmation from
another quarter. Three times in the book of Deuteronomy
are lambs — the characteristic offerings of the passover —
called the '■'■' ashtaroth of the flock," ^ a phrase which prob-
ably had survived from primitive usage, when the connec-
tion of the offering with a deity bearing this name had
been obscured by the introduction of no other epithet.
Further confirmation of this view may be found in the fact
of the sanctity of the threshold which is prominently
recognized in the ritual of the passover. ^ Trumbull ^
traces this sanctity back to a recognition of the relation
of the sexes to one another, and, although my own studies
would lead me to think that the direct application which
he makes of it to the passover leaves out of account some
other important elements, yet his explanation of the fact
harmonizes with the general explanation of this cult which
we have reached. All these phenomena connected with
the passover, therefore, confirm the view that that festival,
even in the worship of Yahwe, goes back to a primitive
Semitic root.

6. The origin of Yahwe for which we contend is con-
firmed by the most probable etymology of his name — that
proposed long ago by Le Clerc * and accepted by many
modern scholars — viz.: that the name Yahwe is a Hiphil
form meaning, "He who causes to be," i.e. "gives life."^

iDeut. 713, 284-18.

2 The name nDS probably has nothing to do with leaping over the
threshold, but seems to mean a " dance." Cf. Toy, in JBL., Vol. XVI,
p. 178 ff., and Buhl, Gesenius'' Handworterhuch, 13th ed.

' The Threshold Covenant, ch. v.

* In his Commentary, on Ex. 6', published in 1696. Cf. Driver in the
Oxford Stndia Biblia, Vol. I, p. 13. Le Clerc made a somewhat different
application of it from that advocated here.

^ The name occurs in the Old Testament in four forms, Tahwe, Yah,
Y6, and Yeh6. The second form occurs in proper names and in late
poetry, and the third and fourth in proper names (cf. Gray, He-
brew Proper Names, p. 149 ff., and Bonk, ZATW., Vol. XI, p. 126 ff.).

YAHWE 283

The explanation given in Exodus 3"*- no doubt repre-
sents the understanding of the name prevalent in Israel

The prevailing opinion among scholars is that the shorter forms are de-
rived from the longer form, Yahwe. Priedrioh Delitzsch, however, Wo
Lag das Paradies, pp. 158-164, held that the shorter forms were the ear-
lier, that they were derived from the name of the Bahylonian god Ba or
Ya, and that the longer form was developed from this by the Hebrews, a
view which has not met with general acceptance (cf. Driver in Studio
Biblia, Vol. I, pp. 4-6, and 10 f£.), although Hommel holds it {Anc. Heb.
Trad., p. 114). Margoliouth has more recently revived it in a crude
form which has, so far as I know, convinced no one (cf. Contemporary
Review, October, 1898, p. 581 fl.). The Babylonian origin is not made out
(cf. Sayce, Higher Criticism and the Verdict of the Monuments, p. 87 f£.,
and Early History of the Hebrews, p. 164 ff.), and it is hardly possible
philologically to derive a long form like Yahwe from a, short form like
Yah. Words everywhere wear down, but are not lengthened. More
recently Spiegelberg {ZDMG., Vol. LIII, p. 633 ft.) has proposed an
Egyptian origin for the name Yahwe. It can, however, hardly be sup-
posed that a people whose religious ideas hardly influenced those of
Israel at all furnished them the name of their God.

Most scholars have sought a meaning for it in Hebrew and have ex-
plained it as follows : —

1. As a Qal of mn an old form equivalent to nTl in the sense of "He
who is," i.e. "the self -existent " or "unchangeable one," following Ex.
3"; so Dillmann, Com. ub. Ex., in loc, Franz Delitzsch Com. ub. &en.
(1872) p. 26, 60, and Oehler {Theol. of 0. T., % 39). This form is, as
has been remarked in the text, too abstract to be primitive.

2. As a Qal in the sense of "He will be," also based on Ex. 3", and
Hos. 2^. This theory of the name has given rise to several different inter-
pretations : Robertson Smith {British and Foreign Evangelical Beview,
1876) explained it as "He will be it," i.e. all that his servants look for;
Driver (Studia Biblia, Vol. I, p. 17), Hommel (Anc. Heb. Trad., p. 114),
Marti, Theologie (3d ed., p. 61, n. 20): "He will approve himself,"
i.e. give evidence of his being, or assert his being, will reveal him-
self, or enact history ; Skipwith : " He will be with us," i.e. in battle (in
Jewish Quarterly Beview, July, 1898). Of the applications of this expla-
nation, the first and second are too abstract to be primitive. The third
would do very well if we could be sure that the god who first bore the
name was conceived chiefly as a god of war. Such a supposition is pre-
carious, for as we have seen the early Semite looked to his god to do
whatever he needed, and the war function was only one of many.

3. As a Hiphil from Hin (cf. Arab ^jJ6 and Job 38^) in the sense of
" cause to fall," i.e. to send down. This explanation has received various
applications, as follows: Robertson Smith (Old Test, in Jewish Church,
1st ed., p. 423) and Barton (Oriental Studies of Or. C. of Bhila., p. 87) :

284 SEMITIC ORIGINS

in the prophetic period, but, as many scholars have felt,
it is too abstract to be primitive. Smend and Piepen-

" He who sends down rain " ; Wellhausen {Heidentum, 1st ed., p. 175) and
Stade {Geschichte, Vol. I, p. 429): "He who causes enemies to fall" ;
Margoliouth (PSBA., Vol. XVII, p. 57 ff.): " He who sends down law" ;
and Holzinger (Ein. in das Sexateuch, p. 204, and Com. ilb. JExodus, p.
13): "He who causes to fall," i.e. the destroying demon or destroyer.
For reasons already explained this etymology now seems unsatisfactory.

4. The etymology suggested by Le Clero has been adopted by several
modern scholars, taking the name as a Hiphil of mri = ri\n. Not all, how-
ever, take it in the same sense. Gesenius (Thesaurus, 1839, p- 577, n.),
Baudessin {Studien, Vol. I, p. 229), Schrader (in Sohenkel's Sibel-Lexi-
con), and Schultz ( Theologie, 2d ed. , p. 487 ft.) take it in the sense of " He
who causes being" or "life"; Kuenen {Religion of Israel, pp. 279, 398),
"He who gives existence"; and Lagarde (ZDMCr., Vol. XXII, p. 331,
Symmicta, Vol. I, p. 104, Psalterium juxta Hebrmos Hieronymi, p. 153
ff., Orienialia, Vol. II, pp. 27-30, and Gott. Gel. Anzeigen, 1885, p. 91)
and Nestle {Isr. Eigennamen, p. 88 ff. ) take it as "He who brings to
pass," i.e. the performer of his promises. Of these the general nature of
Yahwe, which a broad view of Semitic development leads us to take,
makes " He who gives life " the most probable original meaning.

There are some traces of the name Yahwe among non-Israelites which
are interesting. Among these I do not count names ending in Ya, for as
Jastrow has shown {JBL., Vol. XIII, p. 101 ff.), such names do not nec-
essarily contain a divine element. This applies even to Bit-ya, which W.
Max Miiller {Asien und Europa, p. 312 ff.) finds in a list of Thothmes
III. Ya-u-hi-i-di, a king of Hamath in the days of Sargon (see Schrader,
KAT.'^, p. 23, and KB., Vol. II, p. 57), in whose name Tahu appears
as a divine element, is very Interesting. It suggests the possibility
that the Kenites who in earlier days settled in the north had extended
their influence to Hamath, so that the epithet by which they called their
god had been applied by the Hamathites to their Hadad. It is possible,
of course, that the Aramaeans developed the name independently. Bau-
dessin has shown (Siudien, Vol. I, p. 180 ff.) how the name Yahwe
passed to Greek writers from the Jews as Tdoi. Macrobius {Saturnalia,
I, 18, 19 ff.), connects the name 'Idu with the Clarian Apollo. There is
also considerable evidence which was collected by Movers (see Phoenizier,
Vol. I, pp. 542-547), which connects 'Ida with the Phosnician 'ASavu.
Lenormant thought (Lettres assyriologique, 1st ser.. Vol. II, pp. 196-
201), that the Phcenicians also had the name as applied to this god in the
sense of the " self-existent one." Driver (iStudia Biblia, Vol. I, p. 3)
claims with considerable force that the name, if derived from mn can
hardly have been of general Canaanitish usage, because in Phoenician as
in Arabic and Ethiopic the substantive verb in ]'Q. It is possible that
they used mn also as well as their Hebrew and Aramaic neighbors, only

YAHWE 285

bring's ^ objection that Israel did not in the Old Testa-
ment period look upon Yahwe as especially the creator is
wide of the mark, if Yahwe was his name first among the
Kenites. To find its meaning we must look at the reli-
gious conceptions of the Kenites, and not those of later
Israel. The Kenites were without doubt in their general
religious conceptions practically on a level with their
Semitic neighbors of the period, and among such peoples
nothing would be more natural, as the preceding pages
have shown, than to call one of their gods of fertility the
giver of life.

Indeed, there is some evidence to show that the name
was actually employed far beyond the bounds of the Ken-
ites, and that it has entered as an element into at least
one Aramaic proper name.^ Yahwe seems, therefore, to
have been an epithet applied by more than one family of
western Semites to gods of the Semitic life-giving type.

7. Another fact which indicates the connection of
Yahwe with the primitive Semitic cult is the connection
of the Kenites with palm trees. The city of Jericho was
at one time one of their seats,^ and Jericho was a city of
palm trees.* Elim, which was apparently a sacred oasis
in the neighborhood of Sinai, contained its twelve sacred
wells and its seventy palm trees.^ About Sinai itself, in
ancient as in modern^ times, the culture of the palm tree

the word has not chanced to survive in any extant inscriptions. At all
events, the view which we are led to take of the meaning of Yahwe makes
it a tempting hypothesis to suppose that either as a native Phoenician epi-
thet, or as one borrowed from their Hebrew and Kenite neighbors, the
Phoenicians applied the name Yahwe, " the life-giver," to their god of
healing, Eshmun-Adonis, though it may well be that the 'Idw which was
applied to Adonis was of different origin from the'ldu which was borrowed
from the Jews. For a recent account of the occurrence of this name in
Greek sources, cf. Deissmann, Bible Studies, 1901, pp. .321-336.

1 Piepenbring, Theology of the Old Testament, p. 100 ff., and Smend,
Lehrhnch, p. 21, n. 1.

« Yahu-bidi, cf. Schrader, KAT.\ p. 23. Cf. KB., Vol. II, p. 57.

» Jud. 116. 4 Deut. 343. 6 Ex. lb'".

« Cf. Wellsted's Travels in Arabia, Vol. II, p. 12.

286 SEMITIC ORIGINS

must have been known ; and, no doubt, its culture helped
to keep alive among the Kenites the religious conceptions
and practices which their primitive forefathers had con-
nected with that tree. Perhaps the recollection of the
connection of the Kenites with the palm is found in the
story of the union of Tamar^ (Palm) with Judah. If
not the Kenites, the tale at least is evidence for the ab-
sorption in Judah of some clan to which the palm was
sacred, — a clan which seems to have made the palm its
totem. 2 Afterward there was on the border of Judah and
Benjamin a place known as Baal-Tamar, — a name which
bears witness to the worship of a god of the primitive Se-
mitic type. There seems to be some evidence that the
place was once named for the old Semitic goddess ; ^ there
can, therefore, be little doubt that the characteristic Se-
mitic cult was known among the early clans which after-
ward were fused in the tribe of Judah. These clans may
not all have been Kenites, but the union of the Kenites
with such clans, so as to form the tribe of Judah, is itself
proof of affinity between them, and an argument in favor
of the similarity of their conceptions and institutions.

The original connection of Yahwe with the palm tree
also receives some confirmation from the fact that palm
trees formed a part of the ornamentation of his temple as
conceived by Ezekiel * ; and, as Ezekiel is thought to
have had as his model the temple of Solomon, it is prob-
able that they had a place in that temple also. The place
of the palm tree in the book of Enoch ^ may be due to
Babylonian influences ; but, even then, such influences
would be much more readily assimilated if there was a lin-
gering conception that such a tree was fundamentally con-
nected with Yahwe.

1 Gen. 38.

2 Winokler, Geschichte Israels, Vol. II, p. 104, would interpret this
story as the conquest of Judah over the place, Baal-Tamar.

8 Cf. 1 Kgs. 98. Winckler (op cit, p. 97 ff.) is probably right in omit-
ting the conjunction of the Massoretic text, and reading " Baalat-Tamar."
4 Ez. 4118. 6 Eth. Enoch, 24.

YAHWB 287

Our studies, therefore, taken in connection with the
work of critical students of the Old Testament, enable us
to trace the ancestry of Yahwe back to primitive Semitic
times. Primarily Yahwe was not radically different from
other deities of the steppe and the oasis ; and in its ear-
liest form the religion to which Moses introduced Israel
cannot have differed radically from other Semitic cults.
An endeavor will be made a little later to estimate the
content of Mosaism, and to trace the process by which
the distinctly moral elements of the Yahweism of the
prophets were introduced ; but, for a clear understanding
of our subject, it is necessary first to determine something
of the ritual and the religious conceptions which be-
longed to Yahwe in common with other Semitic gods, and
which passed with him from the Kenites to the Israelites.

Critics are agreed ^ that the passover, as distinct from
the feast of unleavened bread, belongs to primitive Yah-
weism.

It is described even by P (Ex. 12"*) as practically a
nomadic festival, — a commensal meal, not unlike those of
Arabic paganism. If, however, our previous investigation
has any bearing on the primitive nature of Yahwe and his
worship, there must have been some sexual conceptions,
and probably in the earlier days some similar rites, con-
nected with the passover, which in P's account have been
eliminated. It is, of course, possible that among the Ken-
ites less stress may have been laid upon these elements
than among the Semitic peoples generally, but such a sup-
position is hazardous and cannot be accepted without clear
proof. In later times we find Hannah at the time of
Yahwe's festival — probably the passover — praying for
offspring and gaining the answer to her prayer,^ a fact
which shows that there still survived in connection with
Yahwe's feast some of those conceptions of fertility which

1 Cf. e.g. W. K. Smith, Bel. of Sem., 2d ed., pp. .333 ff., 346 ft. ; Well-
hausen, Prolegomena, 5th ed., ch. iii; Budde, Bel. of Israel, p. 73 ff.
» 1 Sam. 1.

288 SEMITIC ORIGINS

pertained to the primitive goddess.^ Later Hebrew senti-
ment explained the misfortunes of the house of Eli ^ on
the ground that he did not restrain the loose conduct of
his sons upon such occasions, but it is not impossible that
in Eli's time such license may not under the excitement of
Yahwe's festival have been considered wrong. The anal-
ogy of other Semitic deities would lead us to expect that
in their worship of the giver of life and fertility the Se-
mitic tendency to license, of which the Hebrews had their
part, would find expression among them similar to that
which it found elsewhere.

Budde,^ though remarking that he can not and would not
assert that the worship of Yahwe in ancient times was re-
stricted to this simple annual festival, makes no attempt
to determine what other features it contained. Beyond
the supposition that victories in war were celebrated by
especial worship of the god, he contents himself with the
supposition that the worship of Yahwe was of an extremely
simple nature. We are, however, now in a position to
point out that the god of the Kenites, who inhabited oases
like Elim and Jericho,* and who roamed over the steppe,
would be celebrated in a second festival in the autumn at
the gathering of the date harvest.^ This festival, after the
settlement in Canaan and the acquirement of agricultural
habits of life, was naturally interpreted as the festival of
the grape gathering,^ but in the book of Leviticus, where
archaic practices are frequently preserved, the memory
that the feast had a nomadic origin is perpetuated in the
name " Feast of Booths, "^ — a name which is rightly in-
terpreted as a survival of nomadic life.* At the time of
the date harvest the nomads gather about the oases to lay
in a supply of dates and to worship the god of the date
tree ; their tents would dot the outskirts of the oasis and
form a striking feature of the landscape. The book of

1 See above, p. 110. *Jud. lis. ' Lev. 23M.

2 1 Sam. 2. 6 Above, p. Ill ff. s Lev. 234»-«.
' Bel. of Israel, p. 75, " Ex. 34^2 ; 23W.

YAHWE 289

Leviticus comes to the aid of analogy, therefore, to prove
a second primitive festival of Yahwe. No doubt in later
times the good things, which the grape harvest with its
quickly fermenting grape juice afforded, gave to the agri-
cultural festival a more luxurious and boisterous character
than attached to the nomadic feast which it displaced, but
that the one was merged into the other there can be no
doubt.

This autumn festival was, as we have seen above,^ pre-
ceded by the rite of wailing for Tammuz, — a custom
which, as Robertson Smith pointed out,^ has survived in the
fasting and humiliation which preceded the Hebrew Day
of Atonement, — a day which itself preceded this autumn
festival. The ritual of the Day of Atonement is probably
a survival under a new interpretation of the worship of
Tammuz, or equivalent god, in connection with the wor-
ship of Yahwe, for there is no more reason to suppose that
this was borrowed from the customs of the Canaanitish
Baalim than that the date feast itself was. If, then, the wor-
ship of Tammuz was a part of the primitive cult of Yahwe, as
it was of other Semitic cults, one may naturally ask if the
primitive goddess Ashtart was not also originally connected
with Yahwe. On this point we have no direct evidence.
A number of scholars^ recognize in the wailing for Jeph-
thah's daughter * a survival of the Tammuz wailing. The
story as it has been preserved to us makes it clear that the
wailing was performed, not for a deceased god as at Gebal,*
but for a goddess as at Carthage.^ Whether this cult in
Gilead was directly connected with the Yahwe cult in
early times is exceedingly problematical. It was probably
connected with some local clan cult of the tribe of Gad or
Manasseh. Winckler has with much acuteuess shown ^

1 p. 100. 2 2{gi, of Sem., 2d ed., p. 411 ff., especially p. 414.

8 W. R. Smith, op. cit., p. 416; Moore, "Judges," in Inter. Grit.
Com., p. 305 ; and Winckler, Geschichte Israels, Vol. II, p. 140 ff.
' Jud. ll*". ^ Above, p. 245. « Above, p. 255 ff.

' Geschichte Israels, Vol. II.

290 SEMITIC ORIGINS

that many of the traditions of patriarchal Israel go back
to myths of Tammuz and Ashtoreth. No doubt in his
application of this solution to the period of David and
after, he has applied his key where it is unnecessary,
but many of his suggestions seem exceedingly plausible.
If they are true, this common Semitic mythology was well
known in ancient Israel, and it is surely a gratuitous sup-
position to claim that it was all borrowed from other
sources than the Kenites. Analogy thus leads us to believe
that probably the Yahwe worship of the Kenites contained
an Ashtart. If such was the case, some will be ready to
urge that that is no evidence that such worship was
adopted by Moses. It must be admitted, however, that if
the Kenites associated an Ashtart with Yahwe, Moses and
the Hebrews would inevitably worship her too. Converts
to a new religion are not its reformers, but its blindest devo-
tees. Gratitude to the deity who had delivered them from
Egypt would compel the early Israelites to take the cult
of that god over in toto. For reasons, however, which
will be adduced a little later such a goddess, if connected
with Yahwe, must as a goddess of the steppe have had a
character comparatively mild and consequently innocent
as compared with the Ashtaroth of more bountiful and
luxurious Canaan, or the Ashtart of the mercantile, rich,
and luxurious Sidon. Increasing wealth increased the
evil tendencies of this cult ; thus Ashtart, " the abomina-
tion of the Sidonians," became a byword even among her
Semitic kinsmen.

Along with the two feasts which can be traced to prim-
itive Yahweism, and along with Tammuz and Ashtart, we
must place the pillar (masseba), common to Yahwe with
other Semitic deities, and which continued to represent
him down to the time of Hosea,i and probably till the re-
form of Josiah. Here too must be placed the 'ashera,
which marked the limits of primitive Semitic shrines, and
which were not eliminated from Yahwe's temple till the
1 Gen. 2822, Hos. 3S Deut. 7^, and 2 Kgs. 23".

YAHWE 291

time of Josiah.i If the foregoing argument be valid,
these objects must have been as much a part of the
Yahwe ritual of early days as of that of any Semitic god.

Yah we, the god of the Kenites, then, — probably Yahwe
as Moses knew him, — was a Semitic god of the oasis and
the wilderness, of the type found in the Arabian environ-
ment. He was a god of life in the broad sense of that
term ; the Tammuz wailing was a part of his ritual ;
probably to his myths were attached all those feminine
associations which are implied in the wailing for Tammuz.
This god, because of the nature of the weather in the
region where his people lived, had become associated in
their minds with clouds, storms, and thunder ; because of
their warlike struggles with their neighbors, he was also
regarded as the giver of victory in war. The new cult,
to which Moses introduced Israel, did not, therefore,
differ as much from the worship of their neighbors, or
even from their own former clan cults, as even critical
scholars are wont to suppose.^ The chief and significant
difference, as has often been said, lay in the fact that the
worshippers were bound to the god by covenant and not
by kinship ; ^ but in this difference, as will be pointed
out in more detail below, lay the possibility of all spirit-
ual progress.

Israel, with her new faith, entered soon into a new land
— a land where nature was more benignant than on the
steppe ; where human effort was rewarded with more
abundant harvests, so that to those accustomed to the
poorer life of the wilderness it seemed a " land flowing
with milk and honey." In this land they found Canaan-
itish tribes dwelling, whose gods had originallj^ been gods
of the wilderness, like Yahwe, with a comparatively
simple ritual, but who in their more luxurious environ-
ment had become considerably transformed. The revolt-

1 2 Kgs. 23«- ". 2 Cf. Budde, Bel. of Israel, p. 73 ff.

8 Cf. W. R. Smith, Mel. of Sem., 2d ed., p. 318 fC. ; Piepenbring, TJieol-
ogy of the Old Testament, p. 30 ; and Budde, op cit., p. 35 ff., esp. p. 38.

292 SEMITIC ORIGINS

ing aspects of their worship had become more revolting,
the inequalities among their worshippers much greater.
How inevitable it was that Israel should worship these
deities Budde has depicted with great clearness and force. ^
It was the commingling of their worship with that of
Yahwe which introduced into the latter some elements of
civilization which were much needed, but which had been
lacking in the Yahweism of the desert. Ultimately, too,
these Canaanitish cults proved not only as reagents for
the purification of Israel from the old clan cults, as
Budde supposes,^ but from the baser and grosser elements
inherent in itself. How this came about we shall try to
sketch presently ; but, for the sake of clearness of
thought, it will first be necessary to consider a little more
fully what the moral contents of Mosaism were.

Much effort has been made to maintain the position
which criticism had reached in the time of Ewald, that
the kernel of the Elohistic decalogue,^ which is repeated
in Deuteronomy,* is of Mosaic origin.^ It is of little avail
to point out that the Egyptian book of the dead, which is
older than Moses, contains nearly all the moral require-
ments of the decalogue.^ Possibility of existence does
not demonstrate actual existence ; and the actual exist-
ence of the moral decalogue in the time of Moses seems
to be made practically impossible by the existence of a rit-
ualistic decalogue in J '' which is evidently older than the
moral decalogue of E.* If the Pentateuch contains any

1 Op cit., pp. 42-60. 2 76jU, p. 71. » Ex. 20. * Ch. 5.

6 Cf. DiUman, AUtestamentUehe Theologie, pp. 58, 105, 228, and 426 ff. ;
Kittel, History of Israel, Vol. I, p. 198 ; Robertson, Early Religion of
Israel, p. 70, n. ; Bruce, Apologetics, p. 209 ; and Peters, President's
address before the Society of Biblical Literature and Exegesis, December,
1900. The argument that the author of Deut. 10 must have known the
moral decalogue in J is not convincing.

6 See ch. oxxv of the Book of the Dead in FSB A., Vol. XVII, p. 216 fi.

' Ex. 34.

* Cf. Wellliausen, History of Israel, p. 392 ff. ; Prolegomena, 5th ed.,
p. 400 ff. ; Kuenen, Beligion of Israel, p. 244 ff.; Briggs, Hexateuch,
p. 189 ff. ; and Budde, Hel. of Israel, p. 172, n.

YAHWE 293

decalogue which dates from the time of Moses it must,
accordingly, be the decalogue of J, which reads as fol-
lows : —

1. Thou shalt worship no other god.

2. Thou shalt make thee no molten gods.

3. The feast of unleavened bread thou shalt keep.

4. The firstling of an ass thou shalt redeem with a
lamb. All the firstborn of thy sons thou shalt redeem.

5. None shall appear before me empty.

6. Six days thou shalt work, but on the seventh thou
shalt rest.

7. Thou shalt observe the feast of weeks and of ingath-
ering at the year's end.

8. Thou shalt not offer the blood of my sacriiice with
leavened bread, neither shall the sacrifice of the passover
be left until the morning.

9. The first fruits of thy ground thou shalt bring unto
the house of Yahwe, thy God.

10. Thou shalt not seethe a kid in its mother's milk.

These commands are almost purely ritualistic, and at
first glance betray, perhaps, to the unpractised eye noth-
ing which might not be Mosaic. True, the command to
worship no other god was not kept ; but it is neverthe-
less possible that it may have existed as a prohibition of
the introduction of other gods into Yahwe's proper do-
main. The second command of this decalogue is really
not a prohibition of idols, but only of expensive idols.
In the nomadic life and among the poorer after the settle-
ment in Canaan there were two kinds of idols : " graven
images," made of wood, and "molten gods," cast of silver
and gold.^ Sometimes the latter were of wood overlaid
with gold. What the decalogue of J really prohibits is
the making of these molten gods, i.e. the carrying of lux-
ury and extravagance into the worship of Yahwe. It is
the protest of Spartan simplicity and religious conserva-

1 Cf. Moore, "Judges," in Inter. Crit. Com., p. 375.

294 SEMITIC ORIGINS

tism against wealth and innovations. This command
might, therefore, well be nomadic. The same may be
said of the redemption of the firstlings of men and of
asses ; it is likely that human sacrifices were outgrown,
except upon extraordinary occasions, before the settle-
ment in Canaan, and other reasons may have led to the
exemption of the firstborn of the ass. The command that
none should appear before Yah we empty, i.e. each should
bring a gift or sacrifice of some kind, is as appropriate to
the life of the wilderness as to that of settled Canaan.
The exclusion of leaven from Yahwe's sacrifices, and the
obligation to consume the passover victim before morning,
are both obligations which were felt in the nomadic form
of life.^ The same is true of the prohibition to seethe a
kid in its mother's milk.

A careful examination of some of the remaining com-
mands produces, however, a different impression. The
keeping of the feast of unleavened bread is an agricultu-
ral and not a nomadic regulation. It must have been in-
troduced into the present decalogue after the settlement
in Canaan ; but it is quite possible that it displaced a
command to keep the passover which stood in an earlier
nomadic decalogue. As the feast of unleavened bread
and the passover were merged into one, it would be very
easy for the agricultural name in course of time to dis-
place the nomadic. Similarly, the command to observe
the feast of weeks is an obligation of agricultural and not
of nomadic life. As it stands it is coupled with a com-
mand to observe the feast of " ingathering," or of " taber-
nacles." In the later Hebrew calendar these two feasts
occurred some months apart ; why, then, should they be
here united in one command ? Is it because the command
is but a rewording of an earlier nomadic law expressive
of the obligation to observe the Tammuz wailing and

1 Cf. the sacrifice of the Arabs, witnessed by the son of Nilus, which
was consumed before the sun obscured the morning star. See W. E.
Smith, Bel. of Sem., 2d ed., p. 338.

YAHWE 295

keep the date harvest festival ? Such a theory is not im-
possible and it is certainly attractive. If we take this
view, the substance of nine of these commands may with
plausibility be attributed to Moses.

Of the tenth, the command to keep the seventh day,
the same in the opinion of some scholars cannot be said.
The sabbath seems to Jastrow and Budde to have been of
Babylonian origin, and not a part of the religion of the
steppe. 1 Budde thinks it became an institution of Yah-
weism during those years when Israel was making the
transition from nomadic to agricultural life, and when
Yahwe was being transformed from a god of the oasis
and the steppe to a Palestinian Baal. Perhaps it was
then organized into the form in which we now have it, but
as Toy^ has shown, it probably goes back to a taboo
which is considerably older. Probably, then, this com-
mand has displaced the expression of this early taboo in
an earlier nomadic decalogue. We have now no means
of proving this, though from what has been said of the
other commands, it does not seem improbable.

We conclude, then, that Moses probably summed up the
precepts of the worship of Yahwe in ten "words"; that if
he did so, the decalogue of J has more nearly preserved
them than any other part of the Pentateuch, but that
even the decalogue of J as it now stands has undergone
some changes since the time of Moses.

In close connection with the decalogue there stands in
the Old Testament the ark, called variously " the ark,"
"the ark of Yahwe," ^ "the ark of the covenant of
Yahwe,"* and " the ark of the testimony," ^ which, accord-
ing to a late tradition,^ contained the decalogue written on
tables of stone. This ark seems to have been a box simi-
lar to those which the Egyptians and Babylonians used

1 Cf. Jastrow in American Journal of Theology, "Vol. II, pp. 312-352,
and Budde, Mel. of Israel, p. 66 ft.

2 Cf. JBL., Vol. XVIII, pp. 190-195.

3 In J, E, and Samuel. « In D. ' in p. 6 i Kgs. 8^- "K

296 SEMITIC ORIGINS

for carrying their gods from place to place.^ Among the
Hebrews the ark probably formed a kind of nomadic
temple.^ The fact that in the Judaean source, J, the ark
plays no prominent part, but Yahwe is represented as
dwelling at Sinai, while his angel goes before Israel,^ and
in E, the Ephraimite source, the ark plays a much more
prominent part,* led Wellhausen and Stade to believe ^
that the ark was originally the movable sanctuary of the
Joseph tribes from whence, after the union of the tribes,
it was adopted by the nation. This view has been
adopted by many others.^ As Moses was the deliverer of
the Joseph tribes, it is altogether probable that the ark
was of Mosaic origin, and was a part of the Yahwe ritual
of the time of the wilderness sojourn.

The difficulties with reference to the decalogue and the
several versions in which it exists, have led these scholars
to doubt the accuracy of the tradition that the ark con-
tained a copy of the table of ten words. They have sup-
posed that it contained a sacred stone or aerolite, similar
to the sacred stone in the Qa'aba at Mecca, which was a
kind of fetich. This may be true, but our analysis of the
decalogue of J has shown us how possible it is that a
nomadic decalogue of ritual lay back of J's ten words.
It would be most natural for such a decalogue to be in-
scribed on such a sacred stone. The tradition, therefore,
seems worthy of credence.

1 Wilkinson, Ancient Egyptians, Vol. Ill, p. 289 ; Delitzsoh, Assyrisches
JSajidworterbuch, under elippu, and "Isaiah" in 8B0T., p. 78.

2 Cf. Wellhausen, Prolegomena, 5th ed., p. 46, n., and Heidentum, 2d
ed., p. 215 ; Stade, Geschichte, Vol. I, p. 457 ; Nowack, Archceologie, Vol.
II, p. 3 ff. ; Benzinger, Archceologie, p. 367 ft. ; Winckler, Geschichte
Israels, Vol. I, p. 70 ff. ; Couard in ZATW., Vol. XII, p. 53 ff. ; and Hop-
kins in JAOS., Vol. XX, pp. 303-308.

Ex. 322.

* Even if it be true, as Driver supposes ("Deuteronomy" in Inter.
Crit. Com., p. 118), that J originally described how Moses made the ark,
that "would not affect this conclusion, for by the time of J, as we have
pointed out above, many of the Ephraimitic traditions had become cur-
rent in Judah and are mingled with the Judsean in J's writing.

'' See references in n. 2, above.

YAHWE 297

Couard believes ^ that the ark was carried from Jeru-
salem by the Egyptian king, Shishak, in the time of Reho-
boam. That would adequately explain its disappearance
from the later history. That disappearance would also
give scope to the traditions to substitute without con-
scious violence the ethical decalogue of later times for the
ritualistic decalogue of earlier days, in response to the
advance of the moral consciousness.

Moses then, we may suppose, gave Israel its Yahwe
worship, its ark as a movable temple, and a ritualistic
decalogue. In course of time the nation passed on from
the steppe, and, attracted by the more fertile fields of
Palestine, won its way into Canaan. It was then most
natural for them to give some worship to the Baals.
Later, in the time of David, it was thought that when one
entered upon a new land it was necessary to worship the god
of that land,2 and centuries later than this the Babyloni-
ans whom Sargon imported into Samaria found it neces-
sary to propitiate the god of the new land in which they
found themselves.^ That the Israelites actually wor-
shipped the Baalim Hosea directly testifies.* The worship
of Yahwe as their own tribal god was also maintained, and
in process of time, as Budde has so well depicted,^ Yahwe
became a Baal, — a god of the land. Agricultural festi-
vals, once celebrated to the Baalim, became festivals
of Yahwe, and agricultural functions, once foreign to
him, were now thought to be his.

The proof that Yahwe became a Baal is of various
kinds, as follows : ® 1. Saul and David, both champions
of the worship of Yahwe, gave names to their sons into
which Baal enters as a constituent element, as Ish-Baal,
Meri-Baal, and Baalyada, — names in which critics gener-
ally agree that Baal is an epithet of Yahwe. 2. The
shrines of Baal became in many places, as Bethel,

1 ZATW., Vol. Xll, -p. M. 4Hos. 25.

2 1 Sam. 2619. 6 jigi of Israel, ch. ii.

1 2 Kgs. 172*-S4. « Cf. Budde, op eit., p. 106 ff.

298 SEMITIC ORIGINS

Schechem, and Hebron, shrines of Yahwe. The processes
by which this was accomplished are described to us in
Jud. 6. It resulted from the conquest of Yahwe and
Israel over the Canaanites and the local Baalim. ^ Yahwe
had proven himseK stronger than these gods by conquer-
ing their land and their shrines. Gradually, as he became
associated with their shrines, traditions arose to explain
how he had consecrated them in former days by revealing
himself to patriarchs or heroes there, so that Israel came
to believe that Yahwe was only conquering back that
wliich had been his own. The old rites continued, but
now they were rites of Yahwe. 3. The transformation
of Yahwe's ritual from the simple nomadic to the rich
agricultural type, and its fusion with previously exist-
ing Canaanitish ritual, is another proof that Yahwe be-
came a Baal. To this transformation the prophets bear
direct witness, — Amos declaring that such ritual formed
no part of the wilderness religion,^ and Hosea that Yahwe
was the giver of plenty.^ 4. Another proof that Yahwe
became a Baal, is the fact that the bull became his symbol.
It has been pointed out already* that in agricultural
communities the bull frequently became the symbol of
the deity, who was regarded as the giver of agricultural
plenty. This became true also of Yahwe in Israel.
Jeroboam could say of the bull images at Bethel and Dan,
which from their diminutive size were called "calves,"
" Behold thy god, O Israel, which brought thee up from
the land of Egypt," ^ i.e. "behold Yahwe." Probably
similar images were in the temple of Yahwe at Gilgal.^
In the temple at Jerusalem the bull symbols appeared in
another form ; they there supported the great laver.
From such facts as these it is clear that when Israel con-
quered Canaan, Yahwe became a Baal, — a god of the
land. This accounts for the fact that Naaman took Pales-

1 Budde, op. cit., p. 103 ff. * Above, p. 201 ff.

2 Amos 521 ff-. 6 1 Kgs. 1228.

' Hos. 2*. 6 Amos 5< «-, Hos. 41=, QW, 12".

TAHWE 299

tinian soil to Damascus in order that he might worship
Yahwe there.^ By the time of Elisha, Yahwe was so much
a Baal that he could be worshipped only on Palestinian soil.
For the same reason at a later time the Babylonians, resi-
dent in Samaria, learned the worship of Yahwe, so that as
god of the land he might not send lions upon them.^

In the development of Yahwe into a Baal, his cult, or
rather the conception of him held by his worshippers, gained
something which was necessary before Yahwe could per-
form for the world the lofty service which lay before him,
for it passed from the narrow, tribal type of religion, hos-
tile to culture and civilization into the broader sphere of
a national religion, capable of adapting itself to the pur-
poses of a finer and more civilized life. This transforma-
tion was accomplished, as Budde has pointed out,^ by the
achievement of Israelitish mastery over Palestine and the
united efforts of prophets, priests, and kings. Meantime,
on the outskirts of the nation lingered the Rechabites, a
conservative force, maintaining the nomadic ideal, and pre-
senting a continual protest against what they regarded as
the degenerate tendency which was Baalizing Yahwe.
The part performed by this element of the nation was in
the end quite as necessary as that of their opponents for
the preparation of the Yahwe cvlt for its high service to
mankind.

This transformation of the god of the steppe into the
Baal of a settled community was by no means an experi-
ence peculiar to Yahwe ; it occurred wherever the nomads
of the oasis and the desert passed over into settled agri-
cultural communities. The Baals of Canaan were, as we
have seen,* themselves only gods who, like Yahwe, had
sprung from Semitic nomadic society and had been Baal-
ized a little in advance of him. Yahwe's kinship to them
hastened in his case the Baalizing process.

While this Baalizing of Yahwe was a necessary part of

1 2 Kgs. 5". « Bel. of Israel, p. 77 fi.

2 2 Kgs. 172'- 2s. 4 Above, pp. 147-150.

300 SEMITIC ORIGmS

the preparation for the place he was to hold in the religion
of the human race, for that place he would have been no
more fitted than any other Semitic Baal, if providentially
the Baalizing process had not been checked at the proper
point, and Yahwe forever differentiated in the minds of his
worshippers from these gods. The outward events which
were the occasion of this differentiation were as follows:
In the reign of Ahab the natural assimilation of Yahwe
to Baal was interrupted by the violent introduction of a
foreign influence. Ahab had married a Tyrian princess,
who was of course allowed to bring the worship of her
native gods with her. Being of an ambitious nature, she
prompted her husband to trample upon the popular rights,^
and thereby aroused the sentiment of the people against
her. She seems to have looked with disdain upon the
simpler religious rites of her new and comparatively rustic
home, and to have endeavored to introduce the more ornate
and voluptuous cult of Tyre. Tyre was at the time one
of the world's great emporia ; through its sea-faring mer-
chants the wealth of the nations flowed into it.^ Its
riches had pampered the lusts of its citizens, and had made
the excesses of that Semitic worship, the rites of which
appealed so strongly to the passions of men, as much
worse than the rites of that worship at Samaria as those
of Samaria were worse than those of the wandering tribes
of the steppe. It was this new and sudden excess of
wantonness combined with oppression which aroused the
opposition of the conservatives in Israel. This opposition
was headed by Elijah the Tishbite, from Gilead, a country
of pasture lands where the forms of nomadic life and the
original ritual of the worship of Yahwe were probably
less disturbed by the settled life of Israel than in the more
productive regions west of the Jordan. Accompanying
this new assertion of popular rights and of Yahwe's abhor-
rence of foreign gods and oppressive, debased morals, there
was manifested a new and unique conception of God and
1 1 Kgs. 2 2 Ez. 27, 28.

YAHWE 301

of ethical standards. How far these were manifested in
Elijah himself it is impossible to say ; but his work was
in successive generations taken up by Elisha, Amos, Hosea,
Isaiah, and the great succession of literary prophets down
to the close of the Babylonian exile, and from Amos on-
ward the new moral and monotheistic conception of Yahwe
can be traced. This is not the place in which to sketch
in detail this prophetic struggle ; those who wish to read
it may easily do so in the masterly little treatise of Budde ^
so frequently mentioned already.

So far as the outward features of this struggle were
concerned, it seemed at the start to be a battle between the
nomadic ideal of Yahwe and the excessively voluptuous
Baal of a wealthy Semitic city, — a struggle which ap-
pears perfectly natural, and indeed inevitable. We can-
not, however, follow the story of the conflict far without
perceiving that there were unexpected issues involved in
it, — that unique ethical standards and conceptions of God
were here struggling for expression, standards which
are quite unaccounted for by their environment. From
Amos onward practical monotheism, social justice, and
purity — a justice and purity which are thought to have
their root in the very nature of Yahwe — are proclaimed.

The way for this proclamation had been prepared by
the covenant which Moses had mediated between Yahwe
and his people. A god bound to his people by kinship
could never exert upon his worshippers an influence for
moral elevation which should transcend their inclinations.
Like an Arabic sheik, he might be angry and neglect his
people for a time, but in the last extremity he must help
them, for his position, nay, his very existence, depended
upon that of his kinsmen. With a covenant god all this
was changed. Bound to his people by contract only,
with an independent existence quite apart from them,
he could easily cast off an unfaithful people who refused
to fulfil their part of the covenant. Upon this fact the
1 Beligion of Israel to the Exile.

302 SEMITIC ORIGINS

prophets seized, and from generation to generation urged
it with persistence and force. ^

This fact would have had little significance, however,
but for the new moral and spiritual conception of Yahwe
which they taught along with it. Never in the Semitic
world before had such lofty conceptions of God been pro-
claimed ; never had such ideals of life been urged upon a
people. While these ideals form the burden of the utter-
ance of all the literary prophets, they did not begin with
them ; they had been felt in part for some time in those
prophetic circles in which the J and E documents were
composed, and probably in germ were harbored in the
breast of Elijah. This prophetic conception of Yahwe
aimed to bring back his cult to what the prophets con-
ceived to be its primitive purity. Such in every age has
been the goal of reform, — to establish Mosaism, or
apostolic Christianity, or whatever the primitive form of
the religion in which the reform is working may have
been. We have not yet reached a point of religious cul-
ture, where men generally are willing to work for, or to
accept, a religious ideal which they are not persuaded
is primitive. To consciously strive for an entirely new
ideal is even now a rare phenomenon in religious ac-
tivity. So the prophets labored and struggled, — Amos,
to get rid of feasts which he declared formed no part
of the wilderness religion ; ^ Hosea, to take Israel away
from her Baal lovers back to the wilderness ideal as he
conceived it, of conjugal fidelity to Yahwe ; ^ and subse-
quent prophets take up similar plaints and labor for similar
ends.*

In this connection it will be of help to a clear under-
standing of the work of the prophets and the outward aids
to their success to note four facts : 1. In this long battle
Yahwe was not only differentiated from the Baals and the

iSee e.g. Amos 32- S; Hos. 2; Isa. S'-'; Jer. 31*; Ez. 20; and Isa.

2 Ch. 52«- 26. a Hos. 2. « Cf . e.g. Isa. li^s and Jer. 3.

YAHWB 303

clan cults of the various Israelitish tribes,^ but from his
own original nature. The Yahwe whose ancestry we
have been tracing was, as Paul would say, the Yahwe
according to the flesh ; in the age of the prophets the
Yahwe according to the Spirit appeared in the world.
Yahwe at the close of the prophetic period — Yahwe, the
one God of the world — was as conceived by his followers
a very different being from Yahwe as worshipped by the
Kenites and by Moses. ^ The latter was, as we have seen,
a god of fertility, pleased with such rites as similar gods
of fertility among Semites of a like degree of civilization
were supposed to sanction. He was less gross than Baal
only because the nomadic environment imposed greater
simplicity of life upon his followers. Yahwe as conceived
by the faithful in Israel at the end of the exile was the
God of the world, just and righteous himself, and satisfied
with nothing less in his followers. The conception of
him then held needed but the broadening and deepening
which was to come in part through the contact of his
followers with a larger world in the succeeding centuries,
and in part through the teaching of Jesus Christ, to be-
come the ultimate conception of God for the ages, — meta-
physically perfect,^ morally perfect,* religiously perfect.^

2. In this transformation of Yahwe the absence of
written religious records of the earlier time was a positive
help. If there were written tables of law in the ark, as
we have seen, they probably disappeared in the time of

1 Budde, Bel. of Israel, p. 71 ff.

* If, as we have supposed, Moses conveyed to Israel a brief summary of
ritual, and as many critics have also supposed, as the leader of Israel he
judged causes (or brought them to the sacred lot of Yahwe for adjudica-
tion), it would be most natural for the successive legislation, each code of
which was designed by its promoters to revive what they conceived to be
primitive Mosaism, to be all ascribed to him.

s " God is Spirit," John 42*.

* " God is light," 1 John 1^. Light is used by this writer as equivalent
to moral purity ; darkness is his synonym for evil.

6 " God is love," 1 John 4*- ^^. Love here has lost its old Semitic phys-
ical meaning. It is love as defined in 1 Cor. 13.

304 SEMITIC ORIGINS

Rehoboam. The whole fate of the ritual and the concep-
tion of what Yahwe required were thereafter committed
to tradition. If one came forward from Judah claiming
one ideal as Mosaic, Ephraim, if she possessed a higher
ideal, could claim the authority of her own traditions as
proof of the Mosaic authority of the loftier conception.
When the Deuteronomic law was afterward found in the
temple, there was no authoritative written bar to its recep-
tion, and as that law appealed to the religious conscious-
ness of the prophets of the time,^ it too could be freely
adopted. Thus freedom for advance without unnecessary
friction was afforded. The ghost of the natural Yahwe
could not rise to successfully contest the rights of the
spiritual Yahwe.

3. The endeavor of the prophets to gain a hearing for
their spiritual conceptions of God and their ethical con-
ceptions of life were greatly aided by the outward events
of Israel's history. A series of national disasters, result-
ing in the overthrow of the northern kingdom in 722 B.C.
and of the kingdom of Judah in 586 B.C., gave especial
point to the teachings of the prophets. The better minds
among the people were thus aroused to listen and obey ;
while the obstinate were absorbed, either among the nations
whither they were carried captive, or among the mongrel
Samaritans where they were left.

4. It should be borne in mind that the prophetic en-
deavor of those centuries did not, in one sense, accom-
plish the ideal which at the beginning (or at least early
in the conflict) it had set before itself. In the eighth
century it had high hopes of sweeping away the ritual
altogether ; ^ but the reaction under Manasseh seems to
have convinced the prophetic leaders that the time was
not yet ripe for trusting their spiritual conceptions to the
stormy voyage of the centuries, unprotected by some ark
of legal forms. The Deuteronomic law was then formu-
lated to embody the new conception of the fundamental

1 2 Kings 22W«. 2 Cf. Amos 5=1-25 ; Hos. 66 ; Isa. iwff-.

YAHWE 305

principles of Yahweism in a practical working form. In
this law all sanctuaries but one were abolished ; all out-
ward paraphernalia which might tend in the popular
mind to associate Yahwe with Baal, or even with the
common root from which both had sprung, were rigidly
excluded. Ritual there was to be sure, but ritual robbed,
in so far as it could be, of power to degrade the wor-
shipper. Massebas and 'asheras were swept away, and all
sexual ritual was absolutely prohibited. In the adoption
of this law, however, the older ideal was in some degree
abandoned, and concession made to practical conditions.
In the earlier days the prophets and the priesthood — at
least in the northern kingdom — appear in strong antago-
nism to one another, but in the Deuteronomic reform
they joined hands. A little later, in Jeremiah and Eze-
kiel, members of the Levitical and priestly circles became
prophets. Ezekiel proposed for the post-exilic days a
modification of the Deuteronomic law ; others of the
priestly circles followed in his steps, till by the time of
Ezra and Nehemiah the earlier prophetic standards were
quite reversed, and legal morality had become the ideal —
instead of the free, spiritual morality of the earlier prophets.

This change seems to have been in its turn providential.
The joyous period, when the inspiring voice of contempo-
rary faith could nerve to noble endeavor, had passed
away ; times were at hand which would try men's souls,
— times when an objective ritual for which Israel could
struggle was a necessity, if she were to survive for the
high service which awaited her. This ritual was codified
and accepted, moreover, at a time when the prophetic
ideals of Yahwe had deeply penetrated both people and
priests, so that the new Levitical law, though compiled
from the ancient and sometimes superstitious ^ usages of
the old local sanctuaries, was so purified of most of its
dross that it reflected the new conception of God.

The outline of the genesis and development of Yahwe
1 Cf. Nu. 5"-2i and the sacrifice to Azazel in Lev. 16.

306 SEMITIC ORIGINS

given above may not be attested by evidence sufficient to
commend it to those who are averse to critical study, or
are unaccustomed to the reconstruction of the origins of
civilization by the restoration to their original environ-
ment of fossil customs, born in barbarism, which survive
long after their origin is forgotten. The evidence is,
however, sufficient, I believe, to carry weight with those
who have some familiarity with investigations in primi-
tive religion and of the nature of the evidence which we
have a right to expect.

The results which our discussions have reached are also
most reassuring to the lover of the Old Testament.
Nothing could show more conclusively than the above in-
vestigation does that the moral standards of the prophets
and their conception of God are utterly unaccounted for
by their environment. The tendency, shared by the an-
cestors of the Hebrews in common with other Semites, to
deify the functions whereby physical life was produced,
could give no promise, when judged by the fruits it pro-
duced in other places, of the rich and pure ethical and re-
ligious harvest which it bore in Israel. The primitive
conception of physical fatherhood became after Hosea^
the conception of a moral father with all the high quali-
ties of an unselfish parent raised to an infinite power.
The early conception of a deity who gloried in the pro-
cesses of reproduction, however savagely they were in-
dulged in, was replaced by the conception of Yahwe as a
tender and affectionate Husband who grieved over the in-
continent pollution of Israel, the bride of his choice, — a
Husband whose love was the embodiment of all purity,
whose rule demanded perfect ethical relations between his
sons, and especially between his sons and daughters. If
critical study makes it impossible for us to trace the birth
of these conceptions back to Abraham or Moses, or to ac-
count for them by the supposition that they descended
from heaven amidst the thunders of Sinai, it nevertheless

1 Hos. 111.

YAHWE 307

emphasizes their real inspiration, for it demonstrates on
the one hand that they first took their shape on earth in
human minds, as all spiritual conceptions must, and on
the other that there was nothing in their physical and
social environment which adequately explains them, —
that, after all, the inspiring touch of these prophetic hearts
by the divine Spirit is their only real explanation. We
go back to the rise of Semitic life, we test its nature at
the root, we trace its many -branched trunk through the
various civilizations ; but we find in none of them except
this little Hebrew branch ^ any potency or promise of
spiritual flower or ethical fruit so rich and fair ; we trace
the outward events of the appearance and growth of this
little branch, we find here a favorable condition, there a
providential adversity, but none of these fully account
for the beauty of the branch or the purity of its flower
and fruit. Nothing approaching it in sublimity^ has
without its help been produced in other parts of the
world. We are compelled at the end of our study to con-
fess that "men from God spake, being moved by the
Holy Spirit. "3

It must be remembered, however, that it is not as
strange as it might at first appear to be, that such spir-
itual conceptions should have been grafted upon the
Semitic stock, which has often seemed so sensual ; for as
was pointed out above,* recent investigation is opening
our eyes to the fact that the religious and moral develop-
ment of the race has been closely bound up with father-
hood and motherhood, and that the periods of religious

1 I do not forget the good points of Mohammedanism, but Mohammed
was clearly indebted to Judaism and Christianity for much of his con-
ception of God.

2 Single thinkers in Egypt, Greece, India, and China may have reached
thoughts similar to these, but the sublimity which appears in Israel is
that of a practical monotheism accepted by the whole nation, — men,
women, and children ; the loftiest thoughts of God applied to daily duties
by all.

» 2 Pet. 1^1. * See above, p. 107.

308 SEMITIC ORIGINS

growth in the individual coincide with the periods of
physical preparation for these functions. Religious prog-
ress has always been most marked where the rational
and mystical elements appear in the happiest combination.
Where the rational element predominates, religion be-
comes a cold formality ; where the mystical is in excess,
it becomes fanciful and extravagant, losing real touch
with life. But the mystical has always delighted to
express itself in terms of spiritual matrimony, and is the
purified form of that which the early Semites far back in
the evolution of civilization so grossly expressed.^ With
all its excesses, therefore, we must consider the widespread
Semitic cult as the preparation of a religious soil, in which
the lofty conceptions of God and duty, which appear so
unique in Israel, could take root and produce their fruit.

1 See the paper of de la Grassarie, read in 1900 at the Paris Congress
of Religion. Cf . Bevue de Vhistoire de religion, Vol. XLII, p. 158.

CHAPTER VIII

BEIEP ESTIMATE OP SEMITIC SOCIAL AND EELIGIOUS
INFLUENCE ON THE NON-SEMITIC WORLD

We have now concluded our brief survey of the birth of
Semitic social and religious life, and its various develop-
ments among the Semitic peoples. Before concluding this
imperfect sketch, it will be helpful to briefly indicate the
various points at which the institutions studied in the pre-
ceding pages have touched and influenced the non-Semitic
world. No extended discussion can be attempted here ;
we shall content ourselves with indicating what the influ-
ences have been, and the points at which they have been
felt. To attempt to follow them out in detail would re-
quire the services of many specialists in several different
fields ; but to ignore them entirely would leave upon the
reader an unjust estimate of the value of the institu-
tions we have been studying as contributors to modern
civilization.

From institutions such as these it is obvious that two
circles of influence would radiate. From the barbarous
Semitic institutions, perpetuated by religious conservatism
far into a succeeding and higher civilization, corrupting
and disintegrating influences would surely radiate. From
the lofty and austere morality of the Hebrews of later
times, from the lofty spiritual vision of the prophets, there
have come, on the other hand, some of the best elements of
subsequent civilization.

In attempting this brief estimate, we shall, for obvious
reasons, confine our attention mainly to the world which
lay west of the Semitic territory. The Semites of the

.W9

310 SEMITIC ORIGINS

ancient Babylonian kingdom of Kish must, through their
colonies in Elam,i have exerted an important influence
upon the kingdom of Elam and upon all the neighboring
states which Elam could influence ; but until more of the
inscriptions of Elam have been discovered and we are able
to read the Elamitic language,^ we cannot even reconstruct
the Elamitic civilization, much less tell the influences which
moulded it.

Similarly we might inquire whether after the time of
Cyrus the institutions of the Persian conquerors of Baby-
lon were affected by their contact with the older Semitic
civilization, but no very positive results can at present be
obtained. Cyrus himself speaks, in his well-known cylin-
der inscription,^ as though he had become a worshipper of
Babylonian gods, or at least of Marduk, but it is probable
that in this respect he was simply a statesman who, as a
matter of fact, kept his own native creed.* At all events,
his immediate successors appear by their religious expres-
sions^ to be practically untouched by Semitic influences,
and to have maintained the worship of Ahuramazda in toler-
able purity. Whether Babylonian architecture influenced
Persian architecture, or the Babylonian religious hymns
the later Persian religious literature, are problems for the
Iranian scholar rather than the present writer. They
would not, even if we had their solution, give much help
in the pursuit of the influences we are now trying to trace.

If now we turn to the other extremity of the Babylo-
nian and Assyrian world, we come upon a territory where
our problem, though far from soluble, presents us, even in

1 Cf. Delegation en Perse. Memoires publies sous la direction de M. J.
de Morgan, Tom. II, Textes elamites-semitiques, par V. Scheil, Paris, 1900.

2 For an attempt to read certain Elamitic words, cf. Jensen's "Alt^
und Neuelamitisciies " in ZDMG., Vol. LV (1901), p. 223 ff.

3 KB., Vol. IIP, p. 120 ff.

* Cf. Gray in JAOS., Vol. XXP, p. 179.

' Cf. Bezold's Achimeniden Inschriften ; Assyrian and Babylonian
Literature, Aldine ed., N. Y., 1901, pp. 171-194, and Jackson in JAOS.,
Vol. XXI2, p. 160 ff.

SEMITIC SOCIAL AND KELIGIOUS INFLUENCE 311

the imperfect state of our present knowledge, with an an-
swer which, though somewhat dim in outline, is probably
in general correct. A group of contract tablets from Cap-
padocia, in the eastern part of Asia Minor, written some-
where between 1300 and 1100 b.c.,^ attest the presence of
a strong Assyrian influence in this part of Asia Minor at
that period. These tablets contain proper names into
which the names of the deities Assur and Ishtar enter as
compounds, and make it probable that the Assyrian reli-
gion, as well as Assyrian culture, made itself felt in this
region at that time.^ It is hardly probable that the wave
of Semitic migration represented by these tablets stands
alone. If the Assyrians had not penetrated into this re-
gion at an earlier time than that just indicated, it is prob-
able that the Aramseans had done so. At least, a little
later their influence, Jensen thinks, can be distinctly
traced.^ Thus they had given the region a touch of Se-
mitic influence. How far Semitic influences coming in
this way penetrated the life and moulded the institutions
of the country, it is impossible now to say. It was evidently
considerable. A little later, and possibly at the time of
which we speak, Cilicia and the regions to the westward
seem to have been occupied by the Hittites, whose monu-
ments indicate that they penetrated to the neighborhood
of Cappadocia.* Hittite monuments are found in many
parts of Asia Minor, and Hittite civilization must have
penetrated the country deeply.^ Not until the Hittite
inscriptions are deciphered can we justly estimate how far
Hittite civilization has been influenced by Semitic.

Jensen, who has struck out a new path for the decipher-
ment of Hittite and has probably rightly identified some

1 Cf. Peiser in KB., Vol. IV, p. viii.

2 For the contents of some of them cf. KB., Vol. IV, pp. 50-57.
' Hittiler und Armenier, pp. 170-177.

* See Messerschmidt's "Corpus Inscriptionum Hettiticarum," in the
Mitteilungen der vorderasiatischen Gesellschaft, 1900, I, p. 21, and II,
Tafeln XXVII, XXVIII, and XXIX.

6 See the work of Messerschmidt just cited, passim.

312 SEMITIC ORIGINS

of the signs,^ has shown that these inscriptions probably
date from 1200 to 800 B.C. While it is probable that the
Hittites were in this region considerably earlier than the
time when their written monuments begin, it is also prob-
able that they had felt the influence of the Semitic contact
long before Tiglath-pileser I encountered them about
1100 B.C. in the region of Carchemish.^ Probability re-
ceives in this case some slight confirmation from other
sources. Some of the specimens of their art, like the
statue, discovered by Koldeway,^ of the weather god, shows
positive evidence of the influence of Babylonian and Assyr-
ian art.*

It is not certain, however, that Hittite civilization was
altogether dissimilar to the Semitic. It is true that many
scholars have regarded the Hittites as belonging to the
Turanian or Mongolian family of peoples,^ while Jensen
believes them to be Aryans,® and the ancestors of the
modern Armenians. Jensen's arguments on this point
are, however, too slender to be convincing. Jensen him-
self has pointed out that many of their characters re-
semble in certain characteristics'' Egyptian hieroglyphs,
while Jastrow claims that many of their proper names
found in Assyrian and Egyptian inscriptions are of the
Semitic type.^ Sergi,^ from anthropological evidence, be-

1 In two articles in ZDMO. , Vol. XL VIII, and his Bittiter und Arme-
nier, 1898. For dates see the latter work, pp. 189-216.

2 KB., Vol. I, p. 33.

' Cf. Messerschmidt, op. cit., Tafel I, Nos. 5, 6.

■* Possibly too at a later time Semitic influences directly from Arabia
were felt here. Eamsay (Cities and Bishoprics of Phrygia, p. 91, n. 2)
inclines to accept a suggestion of Robertson Smith's that Leto, the name
of a goddess of this region, is a corruption of Al-Lat.

^ Cf. Wright, The, Empire of the Hittites; Sayce, Baces of the Old Tes-
tament; Conder, The Hittites.

° Op. cit. ' Cf. Hittiter und Armenier, p. 03.

8 Cf. his article " Hittites," § 12, in Encycl. Bib. The point is of com-
paratively small value because the inscriptions use the term Hittite so
loosely that they frequently refer to Semites under this name. Thus Sar-
gon (KB., Vol. II, p. 57) calls an Aramaean king of Hamath a Hittite.

' Mediterranean Mace, p. 144.

SEMITIC SOCIAL AND RELIGIOUS INFLUENCE 313

lieves that the Hittites were an African race of the same
stock as the Libyans or Berbers, and that all Asia Minor
was peopled by this same stock, which he believes were
one in race with the Pelasgians. In the midst of so many
conflicting views one cannot hold any positive opinion
with reference to the origin of the Hittites, though it may
be pardonable to take the opinion of Sergi as a working
hypothesis. If they are a branch of the great North
African race, it is quite possible that the same oasis influ-
ence which produced the Egyptian Isis and the Semitic
Ishtar may have given them a similar goddess. At all
events, whether from native Hittite conceptions, or Se-
mitic influences, or from both,i the Hittites possessed such
a goddess.^ The evidence of this comes not only from their
monuments, but from the evidences of their influence on
Asia Minor. Hittite civilization spread over all Asia
Minor,^ and it is altogether probable that the Phrygian
goddess, known variously as Rhea, Attis, Cybele,* Leto,
and Artemis, is but a later form of this Hittite divinity,
who, whatever her home-born inheritance may have been,
probably had a considerable element of Semitic conception
about her. She dates from a time when the inhabitants
of the country were totemistic and lived in caves, as many
of her shrines were grottos.^ That this goddess was in
nature the same as Ashtart is clear from the fact that she
was an earth goddess of fertility and love, that a feast was
celebrated to her at the time of the vernal equinox, that
the swine was sacred to her, that ceremonies practically
identical with the Tammuz wailing were yearly celebrated

1 Cults of similar nature would assimilate the more readily.

2 Cf. Jensen, Hittiter und Armenier, pp. 157 ff., 166 ff., and Messer-
schmidt, op. cit., Tafel XXVII, B.

3 Cf. Jastrow, " Hittites," § 11, inHncyc. Bib., and the evidence of the
widely scattered inscriptions in Messerschmidt, op. cit.

* Cf. Strabo, X, 3, 12.

6 Cf. Pausanias, X, 32, 3, and Ramsay, Cities and Bishoprics ofPhrygia,
pp. 89 fE., 138 ff.

314 SEMITIC ORIGINS

to her,i and that she is often described by Greek writers
as androgynous,^ as we have seen the Semitic goddess in
various places to be,^ and that like the Semitic goddess, a
god is in many places represented as her son.* The an-
drogynous character indicates what we also learn elsewhere,
that this goddess of Asia Minor, like the great Semitic
deity, had a long career as a goddess in a matriarchal com-
munity,^ before the changing conditions of civilization
transformed her in some places to a male,^ and that at
some points religious sentiment crystallized (or was em-
balmed in literature) while popular conceptions were in a
confused state with reference to her sex. This cult as has
been said, was widely disseminated in Asia Minor.''

While we cannot claim that this cult in Asia Minor was
solely of Semitic origin, it is probable that it was not only
of kindred origin, but also deeply penetrated by Semitic
influences. The cult of Aphrodite-^neas, which flourished
in the Troad, was, as Farnell has pointed out,^ an offshoot
of the cult of this old Phrygian-Hittite goddess. Much
obscurity attaches to the person of iEneas, but Farnell's
conjecture* that he was the mythical founder of a house
of priestly kings who maintained the worship of the god-
dess seems the most satisfactory explanation of it. The

1 Cf. Baudessin, Studien zur semitischen Beligionsgeschichte, Vol. II,
pp. 188 and 203-207 ; also P. Decharme's article " Cybele," in Darmberge
and Saglio's Diet, des ant. grec. et rom., p. 1682.

^ Cf. Pindar, Pyth., II, 127 ; Pausanias, VII, 17, 10 ; and Lucian, de
Syria Dea, § 15.

8 Cf. above, pp. 148 ff., 181 ff., 244, and 254 ; also JAOS., Vol. XXI^
p. 185 ff.

* Cf. Ramsay, Cities and Bishoprics of Phrygia, pp. 130 ft., 133 fl.,
167 ff., 169 fi.

6 Cf . Ramsay, op cit. , pp. 7 ff. and 94 ff.

« Cf. Ramsay, op cit., pp. 7 ff., 52 ff., 167 ff.

' In addition to references given above cf. Herodotus, V, 102, Pausanias,
III, 22, 4 ; Messersclimidt, op cit., I, p. 33 ; and Ramsay's Cities and Bish-
oprics of Phrygia, pp. 51 fi., 89 ff., 130 ff., 133 ff., 138 ff. For Artemis
at Ephesus, cf. Acts 19.

8 Cults of the Greek States, p. 641.

9 Ibid., p. 638. Cf. Strabo, XIII, 1, 53.

SEMITIC SOCIAL AND RELIGIOUS INFLUENCE 315

myths of the wanderings of iEneas are the story of the
diffusion of this cult.^ By means of these myths we may
trace it to Thrace, to Zacynthos, to Buthrotum, to the
southeast coast of Italy, and to Eryx in the Island of Sic-
ily,^ where it met and mingled with waves of influence
direct from Phoenicia.^ On the way to these points it had
planted itself in southern Laconia,* Arcadia,^ and Argos.^
All forms of the myth, however, represent the goal of
JEneas as Italy, and it is certain that the cult was estab-
lished at various points along the Italian shore of the Adri-
atic,^ at Naples,^ and also at Rome.^ At the latter city it
seems to have been unknown in the days of the kings, but
was afterward introduced from the South. In later times
it became a powerful influence, reenf orced as it was by more
recently imported influences from the East, for the cor-
ruption of Roman society and the destruction of the aus-
tere morals of the earlier Roman period.

From Phoenicia waves of migration to the westward
began at an early date, — probably by 1400 B.C. or earlier,
— and wherever the emigrants went, they carried with
them the cult of their native goddess. We have already
followed in part their course through the islands of the
Mediterranean,!" j^^jt tj^gy also made their way to the main-
land of Greece, where settlements were made at several
points, and Phoenician influence was accordingly a factor
in the resulting religion and mythology. ^^ Thus in Greece
two waves of this cult met and mingled, one from Asia
Minor and the region of Semitized Hittite influence, and

1 Cf. the references in Famell, op cit., p. 737 ff. I am indebted to tliis
work for a number of the references given below.

^ For these places of. Dion. Halic. , I, 39-50.

8 See above, p. 252 ff. ' Catullus, XXXVI, 11.

4 Pausanias, III, 22, 11. » C. I. Gr., No. 5796.

6 Ibid., VIII, 12, 9. 8 strabo, V, 2, 6.

« Ibid., II, 21, 1. " Above, p. 252 ff.

" Cf. Famell, op. cit., p. 618 ff., especially, p. 624 ; Robert Brown, Jr.,
Semitic Influence in Hellenic Mijthology ; and Dyer, The Gods of Greece,
pp. 163-173.

316 SEMITIC ORIGINS

the other from Semitic Phoenicia. One of the plac
where Phoenician influence was most directly felt was
Thebes, in Bceotia. The Phoenician influence is not or
attested by the name, Kadmos,^ but Herodotus was i
quainted with a tradition that Kadmos was a Tyriaj
The traditional origin of this worship at Thebes is cc
firmed by the functions of the Aphrodite worshipp
there. She was a goddess of fertility, who presided o\
the relations of the sexes to one another, and was a]
regarded as the mother of Adonis, the wailing for whc
formed a part of her ritual.^

The cult which thus penetrated Greece from two dir<
tions was spread pretty generally over it.* That it \i
not native to the Greeks is very clear. ^ Perhaps the ci
found its way into Greece at a time before the develc
ment of wealth and luxury in Hittite and Semitic Ian
had removed from its peculiar rites the simple innoceu
of early days ; or perhaps early Greek morals were t
pure to be seriously corrupted by these streams frc
abroad. However this may be, if Farnell is to be 1
lieved,^ Aphrodite in the early years of Greek history ^^
little more than the personification of the power of fertili
and love in life, and neither moral nor immoral. In lai
times — after the fourth century B.C. — this was cons:
erably changed. The influence of the hetserse spread
social life ; national pride sank, and the temples of Aph
dite, as the restraints of the earlier time were thro^
aside, became more and more what they had been
Phoenician cities like Tyre and Sidon, and from ma
centres debased Greek life.

By the beginning of the Christian era, then, this ci
which in one way or another had come to be widely so
tered in the Mediterranean countries, had produced
the society of the Roman Empire, especially in the easti

1 From aip, "the east." = Bk. II, 49. = Pausanias, IX, 16, 3

* Farnell, Ciilts of the Greek States, p. 618 ff.

6 Farnell, ibid., p. 619 ft. « Ibid., p. 664 S.

SEMITIC SOCIAL AND KBLIGIODS rNFLtTENCE 317

portion of it, a condition of social and domestic laxity-
analogous to that in ancient Israel against which Elijah
and his successors had protested. At Corinth, for ex-
ample, the sensuality so strongly rebuked by the apostle
Paul 1 is directly traceable to the corrupting influences of
the temple of Aphrodite which overlooked the city.^ The
corruption thus produced by the religious sanction, which
was thrown over practices which were no longer naif and
innocent, must be set down to the disadvantage of the old
Semitic cult. To that extent it is chargeable for human
degradation.

Out of the society of these times there came, however,
an institution for the birth of which the laxity in social
life, produced by the worship described above, is in large
part responsible, concerning which different individuals
will make widely different estimates. Whether monasti-
cism — for it is this to which I refer — has been on the
whole a blessing to the world depends upon the point
of view from which one looks at it. No doubt there
were many forces at work in the society and theology of
the early Church to produce that exaltation of virginity
and celibacy in the first century which culminated in the
formation of the monastic orders of the fourth and subse-
quent centuries;^ but one of those forces — and one which,
I am convinced, was more potent than has often been
supposed — was a reaction from that sensuality, conse-
crated under the name of religion, which was destroying
the society of the civilized world. It is little wonder that
for a time earnest souls should almost couple the matri-
monial state, even in its purity, with heathenism, and
extol celibacy as the only pure and Christian life.

1 1 Cor. 5, 6. 2 pausanias, II, 5, 1.

' Cf. Kirchengeschichte, von K. Miiller, Vol. I, pp. 208-216, Monasticism,
its Ideals and History, by Adolf Harnaoh, translated by C. R. Gillett,
N.Y., 1895, pp. 5-44, The Monastic Life, by T. N. Allies, London, 1896,
chs. i-iii, Christian Monasticism, by I. G. Smith, London, 1892, oh. v,
Monasticism, Ancient and Modern, by F. C. Woodhouse, London, 1896,
chs. i, ii, and SchafiE's History of the Christian Church, Vol. Ill, p. 158 H.

318 SEMITIC ORIGINS

Doubtless in the centuries which have since elapsed
other causes have perpetuated the monastic orders. As a
means of consecrating life to contemplation and service
they have appealed to ardent individuals ; because armies
of men and women, thus unencumbered by the ordinary
ties of domestic life, have been useful to the rulers of the
church for the accomplishment of their various ends, they
have appealed to the hierarchy ; but in them we have with
us to the present hour an institution which is, in part at
least, a monument to the reaction from the influences for
evil of the worst elements of the old Semitic cult.

In the last analysis, however, the powers for good which
the world has derived from Semitic influence outweigh
those which have made for evil. We should never forget
that the three great monotheistic religions of the world,
Judaism, Christianity, and Mohammedanism, have all
sprung from the religious soil which was prepared by the
primitive cult, the origin and history of which we have
been tracing.

The rise in Israel of the sublime conceptions of God and
duty which created Judaism we have already sketched,^
but we have not hitherto noted the beneficent influence
which Judaism exerted, in the centuries immediately
preceding the beginning of our era, upon the GrBeco-
Roman world. Dispersed as the Jews had been after the
time of Alexander the Great, by contact with the world
their conception of their mission was greatly broadened
and exalted. Formerly they had thought that for the
sake of themselves alone they were the favorites of heaven;
now they regarded themselves as divinely sent missionaries
to the world. A propaganda was accordingly inaugurated,
equipped with an extensive literature,^ to win the world
to Judaism. At the time old national faiths were worn

1 Above, pp. 300-305.

2 Cf. Schiirer's History of the Jewish People in the Time of Jesus
Christ, Div. II, Vol. II, p. 220 it. For a briefer sketch, Thatcher's
Apostolic Church, ch. ii.

SEMITIC SOCIAL AND RELIGIOUS INELUENCB 319

out ; philosophy had taught many the irrationality of
former cults ; the moral sense of numbers was turning in
disgust from social corruption protected under the name
of religion. To these Judaism, with its lofty conception
of God and its austere morals, came as a refuge and an in-
spiration, and at this distance we can only guess at its
power for good ; it must have been immense.

Judaism had, nevertheless, its limitations; it was after
all a national faith. Men could obtain its benefits only by
becoming by adoption members of the Jewish race ; there-
fore, soon after the beginning of our era, Christianity
easily succeeded to its mission. The old Semitic cult had
prepared the soil for Judaism ; both had prepared the soil
for the teaching of Jesus Christ. The matchless figure
of the Master is much less explained by his environment
than the monotheism of the Old Testament prophets ; and
yet it was no accident that the seed of his teaching was
sown on a warm, religious, Semitic soil. Nowhere else in
the world had such a soil been so remarkably prepared.
Christianity, freed through the labors of Paul and such as
he from the trammels of Jewish particularism, with the
prophetic idea of God completed and perfected, with its
consciousness of human brotherhood and the absolute
worth of every individual, went forth to conquer, leaven,^
and renovate the ancient world. Such is the imperfection
of human nature that no ideals, when embodied in human
institutions, are always perfectly expressed or altogether
unmingled with baser metal. It has therefore happened
to Christianity, as to every other religion, that much that
should never have been connected with the name of
religion at all has masqueraded under its garb. Not-
withstanding this, the religion of Jesus Christ has exerted
influences for the moral and spiritual elevation of the
world such as have radiated from no other centre, and
which are simply immeasurable. It is the religion of the

1 See e.g. the 'beautiful description of the effects of early Christianity
in the Epistle to Diognetus, oh. v.

320 SEMITIC ORIGINS

best civilization; it is capable of becoming the religion
of mankind ; its dross is not inherent in it and may be
purged away ; its spirit, its ethics, and its ideals are the
hope of the world. Yet Christianity, with all that it has
been, is, and promises to be, traces its ancestry " according
to the flesh " back to the primitive Semitic cult.

Mohammedanism must not be omitted from this esti-
mate. Though born later than Christianity, and deriving
its monotheism from the same source, its birth in the Ara-
bian peninsula, where civilization had reached a less ele-
vated plane, placed it at a great disadvantage, if judged
from the point of view of an ethical civilization. Its
prophet during the earlier years of his career was earnest
and sincere, and the recipient of a genuine inspiration ;
but in later life he departed from this lofty plane, and
lived for ends which were not entirely unselfish and are
not above the suspicion of sensuality. Its book, the
Qur'an, legislates on the plane of the simple and half-
barbarous life of the Arabian desert for the civil and
religious polity of the world for all time.

No doubt Mohammedanism has in many parts exerted
an influence for good. Where its sway has extended
over races of a lower order of civilization than that of
Arabia, it has tended to elevate them ; but it stunts and
blasts higher civilizations wherever it comes in contact
with them. Perhaps when at its best, in the Middle Ages,
it was nearly on a par with the Christianity it opposed ;
but when its best products in the way of civilization
to-day are placed by the best products of Christian civili-
zation, the verdict of superiority does not fall in favor of
Mohammedanism. Traditions have done much to modify
the application of the teachings of its sacred book ; differ-
ences of temperament and casuistry have produced almost
as many sects and varieties of thought as those which
have sought to express Christianity, varying from the lit-
eralism of the Wahabites to the mysticism of the Fati-
mites and the Persian sects, but wherever it is found its

SEMITIC SOCIAL AND RELIGIOUS INELUENCE 321

spirit and ideals fall so far short of the highest that the
best civilization seems impossible under its rule. Its evil
and its good alike possess elements in common with all
human good and all human imperfection, but it has some
imperfections which are peculiarly Semitic. It is a crude
product of the Semitic religious soil ; it is not wanting in
noble elements, but the acids of the earlier stages of the
growth of Semitic religious fruit have not been, as in
Christianity, ripened out of it. However necessary these
acids may be to the flavor of the ripened fruit, it is fatal
to the flavor of that which wilts before it ripens.

An investigation such as that we have been pursuing
makes it very clear that Kenan's ^ hypothesis of a primi-
tive Semitic tendency to monotheism (at least as at first
presented) can no longer be maintained. If in the reli-
gious sphere the Semites have anywhere proven them-
selves worthy teachers of the race, it has not been because
they had at the first a clearer conception of monotheism
than others, but because the circumstances of their desert
and oasis environment led them in their religion to em-
phasize those functions of life which are most closely
connected with the growth of moral and religious feel-
ing in the individual and in the race.^ This emphasis
led them to practices which were in the early time com-
paratively innocent, and which embodied in gross forms
concepts of God which in spiritual form are now the best
religious possessions of our humanity.^ It was thus
slowly, through long ages, as the strata of the earth are

1 Of. Renan's "De la part des peuples s^mitiques " in the Journal
asiatique, 1859, L^histoire generate, des langues semitique, 3d ed., 1863,
p. 5 ft., and History of the People of Israel, Vol. I, chs. iii, iv. What
Renan really claims in the later work is not monotheism, but heno-
theism, — that each trihe had its own god, but did not deny the reality
of the gods of other tribes. This position is a true one ; but the road
from it to monotheism lay through a long development in which tribes
were welded into nations and the tribal deities were formed into poly-
theistic pantheons.

2Cf. above, p. 107 ff.

» I.e. "God is love," of. p. 107.

322 SEMITIC ORIGINS

formed, that by means of this Semitic life and worship a
religious national character was created to which the high-
est conceptions could be intrusted for embodiment in
human life, — in which " the Word could become flesh and
dwell among us." We need here, as always, to remember
that " that is not first which is spiritual, but that which is
natural, and afterward that which is spiritual."

Matriarchates and polyandry have been developed in
many parts of the world,i j^y^; nowhere on such a gigantic
scale as among the Semites ; nowhere else did environ-
ment so long protect the institution and render its effects
so permanent ; no other institution of the kind became
the stock to produce such a noble fruitage ; nor was any
other so situated geographically as to discharge both its
sewage and its nectar into the springs from which the civ-
ilization of our modern life drew its early draughts of in-
spiration. All this seems to have been permitted in the
case of the Semites by a wise Providence, who thus pre-
pared a soil in which the best religious and ethical ideals
could flourish, and who thus brought out of this cult in
the end more of good than of evil.

1 See above, p. 59 ff.

GENERAL INDEX

A.bdi-kbeba, wrote letters from Jeru-
salem, cir. 1400 B.C., 242; complains
goTernment is being OTertbrown,
274.

Abel, Ludwig, 227.

Abu, meaning both "father" and
"husband," 68; an early epithet of
Sin, 201.

Abyssinia, 8, 25, 29; marriage in,
48 ff., 66; sycamore sacred in, 89;
Semitic religion in, 135 ff. ; agricul-
tural nature of, 138.

Adapa myth, 264.

Adar, possibly a name of Ninib, q.v.

Adonis, Greek name of Tammuz-Esh-
mun, an epithet, 86, 263; worship
of, at Gebal, 245 ff., 265; in North
Africa, 256.

.S;iian, 37, 253.

.Xneas, story, a myth of Ashtart cult,
255, 314 ff.

^sculapius, a Greek name for Esh-
mun, 252, 265, 267.

Afar or Dankali, 10, 25.

Africa, held to be home of Semites,
6; of Caucasic race, 7; northern
part separated from southern, 18;
from Europe by end of last glacial
epoch, 19; home of Hamito-Semitic
stock in north of, 23; Semites
crossed into, 29; Baal worship in,
150 fE. ; Ashtart worship in, 253 ff.

Agade, a city of Babylonia, 162 ; noted
for its grain, 158; for its dates, 159;
held hegemony for a time, 163 ; seat
of worship of Shamash, 212; older
than Larsa, 213.

Agriculture, beginnings of, in Pales-
tine, 146; in Babylonia, 166 ff.,
171 ff. ; individual property in land
in Babylonia, cir. 6000 B.C., 158 ft. ;
connection with growth of cities,
162, 171 ff. ; effect of, on decalogue,
294 ff.

Aksum, capital of Semitic kingdom in
Abyssinia, 135.

Alashia, El-Amarna letters from,
thought to be Cyprus, 250.

Al-Fals, an Arabic god derived by
epithet from Athtar, 134.

Al-Galsad, an Arabic god developed
from Athtar, 134.

Alilat, Greek name of Al-Lat, 234.

Allah, God of Islam, developed from
Semitic mother goddess, 131 ; said
to have daughters, 235.

Al-Lat, daughter of Allah, 133 ; mother
of Dhu-'l-Shara, 133 ; worship at
Taif, Petra, and Palmyra, etc.,
233 ff. ; goddess of unwedded love,
234 ; name an epithet, 235.

Allies, T.N., 317.

Al-Uqaisir, an Arabic god developed
from Athtar, 134.

Al-Uzza, lived in samura trees at
Nakhla, 88 ; connected with Meccan
sanctuary, 133, 235 ff. ; daughter of
Allah, ibid.; companion of Al-Lat,
234 ; nature and worship of, 235 ff . ;
an Ishtar (Athtar) , 236 ff. ; mean-
ing of name, 237.

Amiaud, Arthur, 185, 186, 187, 189, 190,
191, 197, 200, 207, 210, 220, 260.

Ammianus Marcellinus, 47, 56, 61.

Amorites, old inhabitants of Canaan,
147.

Amr b. Kulthum, an Arabian poet,
62.

Antarah, an Arabian poet, 56.

Antiphanes, 251.

Anu, father of Ban, 191, 196; locality
of, not known, 195 ; in oldest triad,
206; origin obscure, 218; head of
pantheon after Gudea, 219 ; god of
Der, ibid.; partly an abstraction,
ibid.; originally a chthonic god of
fertility, 220; temple of, in Ashur,
222.

Anu-banini, king of Lulubi, 200; wor-
shipped Enlil and Ninlil, 203; sig-
nificance of name, 220 ; worshipped
Kamman, 225, 229.

323

324

GENERAL INDEX

Aphrodite, Greek name of Ashtart,

249.
Apollo, a name for Baal in Rhodes,

252.

Arabia, cradle of Semites, 4, 24 ff.,
28 ff . ; why Semites entered, 26, 119 ;
had it forests once? 26; date of
Semitic occupation, 27 ; physical
character, 28; low civilization, 32;
oasis life, 33; poor outside oases,
71; no fishing in, 74; some hunting,
ibid.; once better watered? ibid.;
produced early civilization, 76;
early religion of, ch. iii. ; Arabia
Felix, 124; later religion of, 125 ff. ;
survivals of primitive goddess in,
233-237.

Arad-Sin, king of Larsa, 199; wor-
shipped Ishtar of Khallabi, 260.

Aramseans, their god, 224 ff. ; their
goddess, 239 ff. ; in Palestine, 271.

Arbela, a Semitic city in Assyria, 222 ;
worship of Ishtar in, 262.

Ark of Yahwe, 295 ff.

Arnold, Friedrich August, 62.

Arnolt, William Muss-, 84, 111, 113,
209, 222, 223.

Artemis, an earth goddess, 178; of
Hittito-Semitic origin, 313.

Aruru, a name of Ishtar, 257.

Asher, a clan, 31, 32; tribe of the
goddess Ashera, 248 ff. ; a god
Asher, 249; equated with Yahwe,
ibid. ; an Israelitish tribe, 271.

Ashera, a post, 106 ; a goddess, 246 ff . ;
consort of Hadad, 247; in Mesopo-
tamia, ibid.; name derived from
pole, 248; pole at Yahwe shrines,
290 ff; goddess changed to a god,
249.

Ashtar-Chemosh, god of Moab, devel-
oped from Athtar, 141 ff.

Ashtart, Ashtoreth, Ashtaroth Kar-
naim, a trans-Jordanic town, 238.

Ashtart, totemism of, 37; associated
with water god, 87; goddess at
Sidon, Tyre, etc., 148 ff.; symbol-
ized by cow, 201; worshipped at
Ashtaroth Karnaim, 241 ff. ; at
Sidon, 243 ff. ; "name of Baal,"
244; patroness of mariners, ibid.;
identified with moon, ibid.; at
Tyre, ibid.; at Byblos (Gebal),
244 ff. ; carried to Mediterranean
countries, 249 ff.

Ashtoreth, O.T. name of Ashtart, 148.

Ashkelon, seat of worship of Ashtart

and Atargatis, 241 ff.
Ashur, chief god of the Assyrians,

221 ff. ; a transformed Ishtar,

222 ff . ; derivation of name, 223 ;
folk etymology of, 224, n. 3.

Ashur, the old capital of Assyria, 221 ;
a Semitic town, 222.

Asia Minor, Semitic influence in
311.

Assurbanipal, king of Assyria, 154,
217; repaired temple of Ishtar at
Erech, 257; worshipped Ishtar at
Nineveh, 261 ; at Arbela, 262.

Assurnasirpal, king of Assyria, 154 ;
exhibits popularity of Ramman, 226,
n. 2.

Assyria, dominion of the city Ashur,
221.

Astar, chief Semitic deity of Abys-
sinia, 135; worship carried from
Arabia, 136 ff.

Atar, Aramaic name of Ishtar, 239 ff . ;
nature of, 241.

Atargatis, origin debated, 238, ff. ; in
Aramaic, " Atar-'atah," 239; associ-
ated with Hadad, 239, ff. ; Jensen's
theory of, 240; a composite deity,
240, ff. ; nature of, 241, ff ; fish form
of, 242 ff.

Athirat, a Minsean goddess, derived
from 'ashera, 131, 247.

Athtar, Sabsean god of fertility, 86, 87 ;
transformed from mother goddess,
87; called "mother" and "he,"
125 ff; retained features of mother
goddess, 126 ff. ; localized in differ-
ent places, 127 ff. ; developed by epi-
thets into other gods, 128 ff.

Ati, name of a goddess, probably
Attis, 239 ff.

Atonement, Day of, connected with
Tammuz wailing, 114, 289.

Attis, a Phrygian goddess, 240; com-
pounded with Atar in Atargatis,
240 ff., 313.

Augustine, 42, 100, 110, 254, 255.

Ava or Awa, ancient name of Yeha
in Abyssinia, 135.

Baal, name applied to well-watered
land, 105, 127; god of each Palestin-
ian, Phoenician, and North African
locality, 148 ff. ; Baal-Ham man, Baal-
Barith, etc., 148; worshipped on
hilltops, 151; in Rhodes, 253; Baal-

GENEEAL INDEX

325

Hamman in North Airica, 253 ff . ; in
Cyprus, 266.

Babylon, gained hegemony of Baby-
lonia, cir. 2300 B.C., 163 ; connection
with Gishgalla, 207 ff.

Babylonia held to be home of Semites,
1 ff . ; wrongly, 22 ; civilization of,
155 ff. ; nature of religion of, 171 ff.

Baethgen, Friedrich, 141, 144, 239, 240,
265.

Ball, C. J., 86, 160, 186, 207.

Bambyce, also called Hierapolis and
Mabug, 239, 241, 243.

Banks, Edgar James, 205, 228.

Bantu language, not related to Semit-
ic, 17 ; polyandry of Bantu race, 60.

Barras, an Abyssinian god, 138 ff.

Barth, J., 254.

Basques, 18 ; not related in language
to Berbers, 19.

Battersby, G. Harford-, 270, 276, 278.

Bau, goddess of Uruazagga, 185 ; cult
of, 189 ff. ; a Semitic goddess, 190 ff. ;
meaning of name, ibid.; daughter
of Ann, 191, 196.

Baudissin, Graf von, 226, 284, 314.

Baudoin, Jean, 177.

Bedza language, 10.

Belin language, 10.

Belkassen ben Sedira, 10.

Belser, C, 225.

Benjamin, a clan, an offshoot of Jo-
sephites, 271; meaning of name,
ibid., n. 5.

Bent, J. Theodore, 25, 49, 89, 112, 135,
136, 137, 138, 237.

Benzinger, Immanuel, 68, 99, 115, 296.

Berbers, a white race, 16; identical
with Iberian race, etc., 18 ff. ; lan-
guage of, kindred to Egyptian , 20 ;
their independent system of writ-
ing, 20 ff. ; polyandry and date cul-
ture among, 117.

Berger, Philippe, 150, 254.

Berossos, 91.

Bertholet, A., 85.

Bertin, G., 6, 22, 76.

Bezold, Carl, 225, 310.

Bickell, E., 236.

Blandford, W. T., 138.

Bliss, F. J., 137.

Bloch, A., 254.

Blunt, Lady Anne, 55, 71, 75.

Bne-Ebed-Ashera, a clan in the El-
Amarna period, perhaps same as
Asher, 32, 246 ff., 248.

Bokhari, 65, 105.
Bonavia, E., 90, 93.
Bonk, Hugo, 282.
Borelli, Jules, 25.

Borsippa, suburb of Babylon colo-
nized from Shirpurla, 210 ff.
Breasted, J. H., 274.
Briggs, Charles A., 292.
Brinton, Daniel G., 6, 7, 13, 16, 18, 23,

24, 25.
British polyandry, 61; possibility of,

among Semites, 69.
Brockelmann, C, 78.
Brown, Francis, 226.
Brown, Robert, Jr., 37, 116, 315.
Bruce, A. B., 292.
Brugsch, Heinrich, 10.
Briinnow, E. E., 113, 160, 182,192, 193,

194, 196, 201, 207, 211, 215, 218.
Budde, Karl, 89, 95, 108, 148, 149, 203,

272, 275, 276, 277, 279, 281, 287, 288,

291, 292, 295, 297, 298, 299, 301, 303.
Biihler, Georg, 59.
Buhl, Franz, 68, 226, 238, 282.
Bull, as symbol of Athtar and other

gods, 201 ; of Yahwe, 298.
Bunini, an attendant of the god Sha-

mash, 215.
Bur-Sin, king of Ur, 159; worshipped

Ramman, 225.
Byblos, or Gebal, worship of Ashtart

in, 244 ff.

Csesar, Julius, 60.

Camel, helped to destroy the vegeta-
tion of Arabia, 74; domesticated
early, 75.

Cappadocia, Semitic influence in, 311.

Carpenter, J. Estlin, 270, 276, 278.

Carthage, chronology of, 122; Ash-
tart worship in, 255 ff. ; temple of
Eshmun in, 266, 267.

Catullus, 315.

Ceres, a North African name of Ash-
tart, 255.

Chamir language, 10.

Charles, R. H., 89, 95, 121.

Chemosh, chief god of Moab, 141 ff.

Cherubim, personification of winds,
91, 94.

Cheyne, T. K., 36, 241.

Christianity, influence and power of,
319 ff.

Chronology, of southern and western
Semites, 122; of Babylonia and
Assyria, 153, 154 ; grounds of Baby-

326

GENERAL INDEX

Ionian chronology, 155 n.; of He-
brews, 268.

Circumcision, originated in primitive
Islitar worship, 98 fE.; among He-
brews, 99; among Arabs, ibid.;
Arabian ceremony of, 100 ; a prepa-
ration for marriage, 100 ft.; con-
nected with spring festival, 110;
among Hamites, 115; native Ha-
mitic practice, 117 ; in Yahwe wor-
ship, 280 e.

Cities, origin of, in Babylonia, 162.

Civilization, developed in river val-
leys, 155.

Clan, organization, 30; among Se-
mites, 30 ff. ; genesis of, 34 ; eco-
nomic purpose of, 38 ff. ; age of
republican clans in Arabia, 71 ff. ;
two types of Arabian clan, 267 ff.

Clarke, W. N., 82.

Clay, A. T., 162.

Clement of Alexandria, 251, 252.

Coe, G. A., 107.

CoUitz, Hermann, 250.

CoUizza, Giovanni, 10.

Compound deities, late date of, 141,
151, 240 ff.

Conder, C. E., 312.

Cook, Stanley A., 228.

Cope, Edward, 170, 204.

Coptic language, 10.

Cornill, Carl Heinrich, 270.

Couard, Ludwig, 296, 297.

Covenant, religious consequence of
Yahwe's, 291, 301.

Cow, significance of, as symbol of Ish-
tar and Ashtart, 201.

Cradle of Semites, theories of, 1 ft. ;
Arabia, 24 ff.

Crete, Ashtart worship in, 252.

Croll, James, 14, 15, 19.

Crum, W. E., 10.

Cybele, known in North Africa, 255;
in Phrygia, 313.

Cyprus, Semitic Baal in, 151 ff. ; wor-
ship of Ashtart in, 249 ff. ; influence
of Cypriotic worship on Greece, 252.

Cyrus, conquered Babylon, 154; did
he worship Semitic gods ? 310.

Dagon, a Semitic god in East and
West, 229 ; in Assyria and Palestine,
230 ; theories as to origin of, 230 ff . ;
a transformed goddess, 231 ; proba-
bly not Aramaean, 232.

Damkina, goddess, spouse of Ea, 198.

Dan, Israelitish tribe, 273.

Dangin, Fran9ois Thureau, 106, 146,

158, 159, 160, 161, 164, 169, 182, 187,

198, 199, 207, 218, 220, 225, 261.
Dankali or Afar, 10, 20, 25.
Dates, a fruit, 75 ; gathered in Sept.-

Oct., Ill ff. ; those of Agade famed,

159 ; house for storage of, 160.
Davis, John D., 185, 186, 190, 191, 195,

199.
Dawkins, Boyd, 18.
Death, conception of Ufa after, 95 ft.
Decalogue, the Yahwistic, 110; the

Mosaic, 292 ff .
Decharme, P., 314.
De Goeje, M. J., 5, 24, 88.
Deissmann, G. Adolf, 285.
Dekerto, Greek name of Atargatis,

243.
Delitzsch, Franz, 283.
Delitzsch, Friedrich, 97, 103, 160, 163.

164, 165, 166, 205, 209, 215, 217, 222,

223, 224, 226, 230, 283, 296.
Deniker, J., 20.
Derenbourg, Hartwig and Joseph, 124.

125, 126.
Descent, counted through mother,

51 ff. ; transfer of, to paternal line,

66; from gods, 130; Ishtar's, to

lower world, 259.
Dhu-'l-Khalasa, Arabic god, 134.
Dhu-'l-Shara, god worshipped with

Al-Lat by Nabathseans, 233; sou of

Al-Lat, 234; his worship, 263; a

Tammuz or Adonis, 267.
Dido, a name of Tanith, 254 ff.
Dillmann, August, 275, 283, 292.
Dilmun, an island in Persian Gulf,

211.
Diodorus Siculus, 242, 243, 252, 253.
Diognetus, Epistle to, 319.
Dionysios Halic, 315.
Divorce, among Semites, 45 ff. ; among

Hebrews, 45; Babylonians, 45 ff.;

Arabs, 4Bff.
Dog, as name of a sacred prostitute,

188, 251, n. 2.
Doughty, C. M., 17, 28, 32, 39, 47, 51,

72, 74, 75, 77, 87, 88, 99, 100, 101, 109,

110, 111, 114, 136, 233.
Dozy, R., 69.
Driver, S. R., 3, 102, 103, 104, 141, 146,

244, 251, 252, 269, 276, 278, 282, 283,

284, 296.
Drummond, Henry, 107.
Dumuzi , same as Tammuz, 263.

GENERAL INDEX

327

Dumuzizuab, goddess of Kinunir, 211 ;

precursor of Nabu, 212.
Dungi, kiug of Ur, 153; mentions

Nina, 188; Ea, 198; Sin, 200; Ner-

gal, 216 ; worshipped Ishtar of Erecb,

256.
Dyer, Louis, 252, 265, 315.

Ea, a god, pictured as a fish, 91, 196 ff ;
one of Babylonia's principal deities,
195; god of Eridu, 196; a trans-
formed Ishtar, 196 ff. ; god of wis-
dom, 198 ; member of oldest triad,
206; said to be Marduk's father,
209.

Eabani, story of, 43 ff., 83 ff., 93 S.,
96 ff. ; date of, 44 ; hair like grain,
218.

Eannadu I, (Eannatum), king of
Shirpurla, 153; Eannadu II, Patesi
of Shirpurla, ibid., and 161; con-
quered Elam, 180; mentions Ishtar,
182: Nana, 186; Nina, 188; Ea,198;
ascribes his victory to Enlil, 206;
conquered Gishgalla, 207; wor-
shipped Shamash, 214.

Ebers, Georg, 101.

Economy, cause of paternal kinship,
72; effect on religious conceptions,
82 ; causes transformation of social
structure, 120; economic condition
of south Arabia, 124 ; economic test
of deities, 179 ff ; its value, 221 ;
economic transformation of Yahwe
worship, 291 ff.

Eden, meaning of Biblical narrative
of , 43 ff . , 93 ff . ; origin and develop-
ment of idea, 96 n. ; meaning of ex-
pulsion from, 97 ff.

Edom, the country, 32; name of Baal
among Edomites, 152; Al-Lat wor-
shipped in, 233 ff.

Egypt, language of, 10; did not con-
quer Berbers, 20; nor influence
Babylonian religious conceptions,
169.

Elam, conquered by Eannadu, 180;
conquered south Babylonia, 257;
Semitic influence on, 310.

Elijah, work of, 300.

Enki (Ea), a principal Babylonian
god, 195.

Enlil (Bel), god of Nippur and chief
deity of Babylonia, 163, 195; called
king of countries, 183; superior to
gods of Shirpurla, 185; father of

Sin, 202; originally a Snmerian
god, 204; Semitic element in, 204 ff. ;
member of oldest triad, 206.

Enshagkushanna, lord of Sumir, 153;
devoted spoil to Enlil, 202, 206.

Entemena, Patesi of Shirpurla, 153,
160; built house for storage of
dates, 160; mentions Nana, 186;
Nina, 188; Ea, 198.

Enzu, a name of the god Sin, 199;
meaning of, 201.

Ephraem, the Syrian, 42, 100, 110, 234.

Ephraim, a Josephite clan, 271.

Epiphanius, 233, 234, 263.

Epping, J., 106.

Erech, a city of Babylonia, 162 ; Lu-
galzaggisj, its king, 146, 153; held
Babylonian hegemony at various
times, 163; Ishtar cult in, 256 ff.;
form of, 259.

Eridu, one of the oldest Babylonian
cities, 162 ; held hegemony in Baby-
lonia at various times, 163; oldest
Semitic settlement, 196 ff . ; seat of
a prehistoric empire, 198.

Erim, a section of Shirpurla, 183, 185;
wrongly called Gishgalla, 185, 207;
shown by its goddess to be Semitic,
186 ff., 192 ; a colony of Ur, 200.

Erman, Adolf, 8, 9, 10, 20, 25, 26.

Erua, another name of Dumuzizuab
or Tashmit, 212, n. 4.

Eryx, seat of Ashtart worship in
Sicily, 252.

Eshmun, a name of Tammuz, 92; at
Tyre, 244; reasons for identification
with Tammuz, 265 ff. ; possible ety-
mology of, 267, n. 2.

Eshmunazer II, king of Sidon, 122;
priest of Ashtart, 243; built her
temple, 244 n.

Eth-Baal, king of Tyre and priest of
Ashtart, 244.

Ethiopic language, 10.

Euphrates, overflow of, 156 ff.

Eusebius, 37, 150, 230, 238, 244.

Euting, J., 69, 70, 75, 77.

Ezana, royal author of inscription
from Aksum, 138.

Family, Semitic, 39 ff. ; of primitive
man, 41 ff. ; effect of temporary
marriage upon, 49 ff.

Farnell, Lewis Richard, 178, 255, 314,
315, 316.

Father, not known in Thibetan poly-

328

GENERAIi INDEX

andry, 68; head of Arabian family
by time of prophet, 123; Athtar a
father, 124 ; father and mother com-
bined in one deity, 205; spiritual
conception of fatherhood of Yahwe,
306 ff.

Feasts, number and character of,
108 S. ; in Nisan, 108 ff . ; sacrifices
and lewd ceremonies at, 110; date
festival in autumn. 111; feast of
Mascal in Abyssinia, 112, 137 ; feast
of Tammuz, 112 ft.; agricultural
feasts in Babylonia and Palestine,
114, 288 ff. ; two feasts primitive,
115; feast of Bau, 190; of Adonis
at Gebal, 245 ff. ; spring feast at
Paphos, 251; of Tanith in North
Africa, 254; of Yahwe, 281 ff., 287 ff.

Fell, W., 86, 87, 128, 129.

Fischer, Theobald, 75, 76, 79, 98, 161,
171.

Fishing, no important part of Arabian
life, 74; at Eridu, 196 ff.; at Ashke-
lon, 242 ff.

Fiske, John, 14, 18, 107.

Frazer, J. G., 85, 114, 179.

Frey, J., 120.

Frost, John, 173, 175.

Gabelenz, Graf von der, 18, 19.

Gad, a clan, 271; in east-Jordanic
country, 273.

Gad, god of the tribe Gad, 249 ; Tam-
muz wailing in cult of, 289.

Galalama, Patesi of Shirpurla, 190.

Galla language, 10.

Gallas, the, 25, 115.

Game, in Arabia, 74.

Gamil-Sin, king of Ur, 159.

Gatumdug, an epithet of Bau,. some-
times treated as a separate goddess,
191 ; once an epithet of Nana, 191, n.
6 and 260.

Gebal, a Phoenician city and seat of
Ashtart worship, 244.

Geiger, Abraham, 132.

Garland, A. A., 6, 12, 13, 15, 16, 25.

Gesenius, Wilhelm, 36, 226, 282, 284.

Ghabghab, a rivulet or trench at
Mecca, 235.

Ghillany, R., 275.

Giddings, Franklin H., 13, 14, 33, 34,
40, 41, 49, 50, 53, 55, 60, 63, 73,
77.

Gillett, C. R., 317.

Girsu, one of the districts of Shirpurla,

185; a Semitic settlement, 192, 194;
name interpreted as " body lance,"
192, n. 1 ; oldest settlement of Shir-
purla, 194 ; seat of original kingdom,
ibid. ; contained a district Kinunir,
210 ff.

Gishban, an old Babylonian town
conquered by Eannadu, 206; situa-
tion and religion of, 218.

Gishgalla, a south Babylonian town
of which Babylon was possibly a
colony, 207 ff.

Glacial epoch, causes of, 14.

Glaser, Eduard, 64, 124, 125, 135, 136,
227.

Goat, destructive of Arabian vegeta-
tion, 74; domesticated early, 75.

Goddesses, the earliest Semitic deities,
120, 179 ff. ; natural agricultural
deities, 199; transformed by war,
180 ff.

Golenischeff, M., 124.

Gottheil, Richard, J. H., 239.

Grassarie, M. de la, 308.

Gray, G. Buchanan, 282.

Gray, Louis, 310.

Greece, goddesses of, 178; Semitic in-
fluence upon, 315 ff. ; corruption of
society in, 316.

Griineisen, Carl, 95, 121.

Gudea, Patesi of Shirpurla, 153 ; sacri-
fices offered to his statue, 168 ; wor-
shipped Nana, 183, 184, 186; Nina,
188; Bau, 189, 190, 192, 199; called
Shirpurla a city, 207; mentions
Khallabi, 260.

Gudua, another name for Kutu (Ku-
tha),215.

Guidi, Ignazio, 1, 2, 22, 75, n. 4.

Gula, an epithet of Ban, became god-
dess of healing, 191.

Gummere, F. B., 179.

Gunkel, Hermann, 98.

Guthe, Hermann, 270, 275.

Guti, a god of the Guti, 217 ff.

Guti or Suti, a people, 37, 199; their
religion, 217.

Guyard, Stanislaus, 164.

Haarbriicker, Theodor, 99.

Hadad, Aramaean equivalent of Ram-
man, 224; equated with R., 226;
worshipped in Damascus, ibid. ;
name means "thunderer," ibid.;
calledRimmon, 227; diffusion of wor-
ship, ibid.; a god of fertility, 228; a

GENERAL INDEX

329

transformed Ishtar, 229; Atargatis

worshipped with him, 239 ff.
Haeckel, Ernst, 13, 16.
Hal^vy, Joseph, 31, 64, 86, 130, 131, 135,

164, 165, 209, 227.
Hamilton, an English traveller, 48.
Hamites, language of, kindred to

Semitic, 9 ff.; blonds among, 17;

form one stock with Semites, 20;

records of, go back to dawn of his-
tory, ibid. ; home in North Africa,

21 ; totemistic, 37 ; passed savagery

when Semites separated from, 73;

polyandrous, 74 ; institutions of, 115

ff. ; civilization derived from oases,

118.
Harding, E. E., 108.
Harith, an Arabian poet, 56.
Harnack, Adolf, 317.
. Harran, a seat of the worship of the

moon god, 202.
Hasa, a region in Arabia, 57.
Haupt, Paul, 36, 43, 83, 85, 91, 96, 97,

103, 113, 165, 168, 183, 184, 200, 214,

218, 257, 258, 259.
Heber, a clan of Asher, 31, 274.
Hebrews, beginnings of the nation,

270 ff. ; proposed identification with

the Khabiri, 274.
Hegra, an Arabian seat of the worship

of Al-Lat, 234.
Hehn, Victor, 75, 76, 161.
Heracles, a Greek name of Baal, 265 ff.
Herodotus, 42, 43, 99, 100, 101, 117,

210, 234, 242, 252, 258, 263, 314, 316.
Heuzey, Leon, editor of De Sarzec's

D^couvertes en Chaldie, see De

Sarzec.
Hierapolis, also called Bambyce and

Mabug, 239, 241, 243.
Hilderbrand, Richard, 59, 60, 72, 73,

77, 119, 146.
Hilprecht, Hermann, V., 146, 162, 166,

180, 181, 197, 198, 200, 202, 203, 204,

213, 217, 219, 220, 225.
Hiram, king of Tyre, 122, 244.
Hittites, in northern Syria before

Egyptian conquest, 147 ; civilization

and language of, 311 ff. ; Semitic

influence on, 312 ff. ; supposed ori-
gin, 313.
Hoffman, E. A., archaic tablet of,
158 ff. ; 161, 213; translation of, 213,
n. 5.
Hoffmann, Georg, 104, 141, 226, 248,
253, 254.

Holmes, William H., 14.

Holzinger, H., 275, 285.

Hommel, Fritz, 1, 3, 21, 22, 76, 106,

124, 131, 185, 207, 208, 211, 223, 247,

249, 274, 283.
Hopkins, E. Washburn, 59, 296.
Hubal, a pre-Islamic name of Allah,

134.
Hull, Edward, 139.
Human life, where first appeared,

12 ff.
Hunting in Arabia, 74.
Hurgronje, C. Snouck, 48, 100, 236.
Husband, living in wife's tribe, 54 ff.

Ibu-al-Kalbi, 235.

Ibn-Hisham, 88, 235, 236.

Ibn-Kutaiba, 233.

Idalion, a seat of Ashtart worship in
Cyprus, 266.

Ihering, Rudolf von, 171.

Il-Azza, a Sabbaean god, 237, n. 4.

Iliad, 249, 250.

Ilmaqqabu, a form of Athtar, 87,
128 ff. ; god of 'Amran, 128.

Immeru, a name of Ramman, 225.

Immortality, Semitic ideas of, 95 ff.

Imr-ul-Kais, 62.

lolaos, a Greek name of Eshmun,
265 ff.

Isaac of Antioch, 236.

Isaf, a Meccan god, absorbed in Allah,
133.

Ishmi-Dagan, a king of Isin, 257.

Ishtar, totemism of, 37; cult of, in
Semitic world, 42; connection with
Eabani, 43 ; cult found, 83 ; a
survival from primitive conditions,
84; impure priestesses of, ibid.;
connection with desire, ibid.; a
water goddess, 86, 92 ff. ; connected
with the palm, 92 ff., 98; with cir-
cumcision, 98; etymology of, 102 ff. ;
goddess of oases, 105 ff. ; at Kish,
181 ff. ; at Susa, 184 ff. ; goddess
of Erim, 185 ff.; of Khallabi, 190,
260 ff. ; symbolized by cow, 201 ; at
Agade, 213; maintained identity
there, 215; among the Guti, 217 ff. ;
coupled with Ashur in Assyria,
221 ff., 224; survivals of, chap, vi;
at Erech, 256 ff. ; daughter of Anu,
259.

Isin, a Babylonian city which for a
time held the hegemony, 163, 257.

Isis, Egyptian earth goddess, 116.

330

GENERAL INDEX

Islam, heathen basis of, 131 ff. ; influ-
ence of, 320 ff.

Israel, beginnings of the nation,
270 ff , ; early disorganized condi-
tion, 273 ; stele of, 274 ff.

Issachar, a clan, 271.

Jackson, A. V. W., 310.
Jacob-el, a place in Palestine, 274 ff.
Jacobs, Mr., 35, 36.
Jacut, see Yaqut.

Jastrow, Morris, 7, 23, 37, 43, 44, 83,
86, 91, 93, 97, 98, 109, 111, 112, 156,

163, 190, 191, 195, 199, 200, 203, 209,
212, 213, 215, 216, 217, 219, 220, 223,
224, 225, 226, 228, 231, 232, 257, 258,

259, 264, 272, 274, 284, 295, 312, 313.
Jehovah, see Yahwe.

Jensen, Peter, 43, 83 ff., 85, 103, 106,
183, 185, 187, 197, 209, 210, 211, 212,
216, 217, 224, 230, 231, 240, 246, 259,

260, 264, 310, 311, 312, 313.
Jeremias, Alfred, 43, 84, 85, 95, 112,

164, 195, 264, 266.
Jerome, 144, 234, 238.

Jesus Christ, effect of His teaching on
the conception of God, 303.

Jethro, priest of Yahwe in Midian,
272; initiated Moses into Yahwe
cult, 276 ff.

Jinn, as partners with God, 36.

Johnson, F. E., 62.

Jolly, Julius, 59.

Joseph, a clan, 271; In Egypt, 272;
delivered by Yahwe, ibid. ; entered
Palestine, 273; name as name of
Palestinian place, 274 ff. ; Moses
mediated covenant for, 276.

Josephus, Flavius, 37, 99, 230, 244.

Judah, a tribe, 271 ; Aramaic element
in, 272, n. 4, 273; absorbed the Ken-
ites, 277 ff. ; knowledge of Yahwe
immemorial among, 278; union of
Tamar with, 286.

Judaism, influence of, 318.

Justin, 255.

Kabyle language, 10.

Karaiudash, king of Babylon, 257.

Karnaim, shortened name of Ashta-
roth Karnaim, 238.

Keane, A. H., 7, 13, 15, 16, 18.

Keasbey, Lindley M., 30, 33, 39, 280.

Kengi, a name of Sumir, 202.

Kenites, a Sinaitio clan, 272; incor-
porated in Judah, 273, 277 ff. ; cham-

pions of Yahwe, 277 ; connected with
Midianites, 277; David said to be
descended from, 278, n. 2; perhaps
extended influence to Hamath, 284.

Kent, Charles Foster, 270.

Kiabiri in the El-Amarna letters, 274.

Khallabi, a colony of Erim, 199; seat
of the worship of Ishtar, 260 ff .

Khammurabi, king of Babylon, 153,
165, 198; worshipper of Marduk,
209; suppressed worship of Nabu,
212; repaired temple of Shamash at
Larsa, 212; at Agade, 214; wor-
shipped Bamman, 225; Dagon, 229;
knew goddess Ashera, 247; wor-
shipped Ishtar of Khallabi, 260.

Khumbaba, god of Elam, genesis of,
184, n. 3.

King, L. W., 202, 209, 212, 213, 223, 247.

Kinnir or Kiununir, a part of Glrsu
and an old name of Borsippa, 210 ff .

Kinship, method of reckoning, 50 ff. ;
through females, 51 ff. ; of Joseph's
sons, 52; of Abraham and Sarah,
ibid. ; among Mandseans, Babyloni-
ans, and Hebrews, 53; of Jacob's
children, 54; through father, 71 ff. ;
in Moab, 140 ff.; in Palestine, 146;
in Babylonia, 159.

Kish, a Babylonian kingdom, 180 ; re-
ligion of, 181 ff . ; conquered by Ean-
nadu, 186 ff. ; by Enshagkushanna,
202 ; colonies in Elam, 310.

Kition, seat of worship of Ashtart in
Cyprus, 250 ff . ; Eshmun worshipped
at, 266.

Kittel, Rudolf, 93, 275, 292.

Koraish, a Meccan tribe to which
Mohammed belonged, 235.

Kremer, von, 1, 2, 22, 75, n. 4.

Kuenen, Abraham, 270, 276, 278, 284,
292.

Kutha (Kutu), an old Babylonian
city, 102 ; Nergal, deity of, 215.

Labid, an Arabian poet, 56.

Lagash, a name of Shirpurla, 184.

Lake dwellers of Switzerland, 16, 17.

Lane, Edward William, 47, 104, 105.

Larsa, a Babylonian city, held hege-
mony for a time, 163, 199 ; a seat of
Shamash worship, 212 ff.

Lasirab, king of Guti, 199.

Law, genesis of Pentateuchal, 303 ff.

Le Clerc, 282, 284.

Legarde, Paul de la, 104, 243, 267, 284.

GENERAL INDEX

331

Lehmann, C. F., 155, 156, 165, 168.

Lenormaut, Fran9ois, 86, 92, 196, 197,
284.

Lepsias, the Egyptologist, 25.

Leto, a name of Attis, 313.

Lfitourneau, Ch. , 66, 67.

Leuba, James H., 107.

Levi, a tribe, 271 ; tlieory of origin,
272.

Levirate, connection with polyandry,
66 ff.

Libit-Ishtar, king of Isin, 256.

Lidzbarski, Mark, 203, 228, 239, 255,
272.

Lubbock, Sir John, 15, 40.

Lucian, 43, 85, 86, 100, 112, 113, 239,
240, 241, 244, 245, 265, 314.

Lugal-Erim, name of a masculinized
Ishtar at Shirpnrla, 183, 187.

Lugaltarsi, a king of Kish, 153; in-
scription of, 181, 187, n. 2, 205.

Lugalzaggisi, king of Erech, 153;
reached Mediterranean, 146, 271;
tendency to deification of, 168 ; men-
tioned Nana, 186; Ea, 198; son
reared by Nidaba, 218; worshipped
Ishtar, 256 ff.

Lulubi, an old Babylonian state, 200,
203, 220.

Lydus, Johannes, 109, 250, 251.

Mabug, a town in Lebanon otherwise

called Hierapolis and Bambyce, q.v.
McAllister, Alexander, 99, 101.
McCurdy, J. F., 163, 164, 165.
McGee, W. J., 170, n 4.
McLennan, John F., 39, 40, 41, 69, 60,

66,67.
Macrobius, 178, 239, 240, 241, 262, 265,

284.
Ma' in, an ancient city and kingdon in

south Arabia, 124 ; gods of, 130 ff. ;

influence on Keuites, 272.
Malik, an attendant of the god Sha^

mash, 215.
Malkatu, consort of Shamash of Agade,

214.
Malkiel, a clan, 31.
Mama, an epithet of Bau, 191, n. 1.
Man, first habitat of, 12 ff. ; antiquity

of, 14 ff. ; effect of natural selection

on mind of, 15 ff.
Manasseh, a Josephite clan, 271.
Manishtu-irba, king of Kish, 184.
Manutu, a Nabathsean god, 233.
Marduk, god of Babylon who displaced

Enlil, 163, 207, 209; a Semitic god,
208 ff. ; absorbed myths of Nabu,
209 ; name derived from sun, 209.

Margoliouth, G., 283, 284.

Marriage, Semitic, 41 ff. ; temporary,
45 ff; in Sunan, 48; in Mecca, 48;
in Abyssinia, 48 ff. ; residence of
wife during, 50 fl. ; of Abraham, 52 ;
of Amnon and Tamar, 52 ; of Tab-
nith, 52 ; interruption of, 55 ; beena
marriage, 55; in Ma'allak&t poems,
56; of Samson, 56; mot'a marriage,
61 ; rise of endogamy, 62 ff. ; for
certain days of the week, 63; by
capture, 71 ff.

Marti, Karl, 283.

Martial, 251.

Maspe'ro, G., 38, 74, 115, 116, 117.

Massaba, phallic symbol of deity, 102 ;
found in Abyssinia, 136; altars at
bases of, 137 ; a part of Yahwe cult,
290 ff.

Mayer, Brantz, 173, 175.

Mecca, marriage in, 48; change of
heathenism to Islam at, 132 ff.

Mediterranean, changes in level of, 17 ;
race of, 18 ; shore of, home of Semitic
stock, 21, 24.

Medr, an Abyssinian god, 138 ff.

Meissner, Bruno, 68, 158.

Melqart, god of Tyre, 149 ; developed
from Ashtart, 150, 244; Eshmun
called, 267.

Menephtah, stele of, 274 ff.

Merx, Victor, 58.

Mesha, king of Moab, 140, 143.

Mesopotamia, held to be cradle of
Semites, 3 ff . ; civilization of, 155 ff.

Messerschmidt, L., 311, 312, 313, 314.

Mexico, civilization and religion of,
172 ff.

Meyer, Ed., 147, 163.

Midianites, connected with Kenites,
277.

Migne, Jacques Paul, 237.

Mill, Hugh Robert, 118, 155.

Moab, physical conditions of, 139; a
land of pastures, 139 ff. ; language
of kindred to Hebrew, 140; society
of patriarchal, 140 ff.; religion of,
141 ff.

Mohammed, introduced into Islam
many spiritual elements, 131;
erected them on substratum of hea-
thenism, 1.32 ff.

Mohammedanism, influence of, 320 ff.

332

GENERAL INDEX

Monastlcism, a reaction from Semitic
corruption, 317 ff.

Monogamy, temporary, in early times,
45 ff.

Monotheism, development of, 307 ff. ;
not primitive, 321.

Monro, Robert, 16.

Moore, George F., 89, 103, 141, 143,
148, 230, 238, 247, 248, 289, 293.

Mordtmaun, J. H., 124, 126, 127, 128,
130, 138.

Morgan, J. de, 8, 20, 26, 310.

Mosaism, moral content of, 292 ff.

Moses, circumcision of , 99, 280; flight
from Egypt, 272 ; mediated covenant
with Yahwe, 275 ; a zealot for primi-
tive Yah we worship, 290 ; connection
with Jewish law, 303 ff.

Mother, descent reckoned through,
51 ff. ; mother goddess the primitive
Semitic deity, ch. iii; transformed
into masculine deities, chs. iv, v; in
Arabia, Palestine and Africa, ch. iv ;
in Babylonia and Assyria, ch. v;
not lost in transformation, 129 ff.
and ch. vi ; mother and father com-
bined in one deity, 205.

Moulton, W. J., 281.

Movers, F. C, 284.

Mugheir, modern name of TJr, 199,
201.

MuUer, D. H., 124, 129, 135, 136, 138,
139.

Miiller, F. Max, 116, 227.

Miiller, Friedrich, 10, 11, 12, 17, 24.

Miiller, Karl, 317.

Muller, W. Max, 8, 9, 118, 146, 231,
237, 242, 248, 274, 284.

Mysticism, sexual aspects of, 308.

Myths, as a clew to origins, 195 ff.

Nabathaeans, an Arabic tribe, 263.
Nabonidos, neo-Babylonian king, 154 ;

worshipped Shamash at Agade, 216.
Nabu, god of Borsippa, 210 ; a Semitic

god, 211 ; said to be son of Ea, 212.
Nabu-apal-iddin, king of Babylon, 154,

215.
Naila, a Meccan goddess, spouse of

Isaf , 133.
Nairs, 40, 69 ff. ; Nair polyandry, 61,

63 ff. ; relation to Thibetan, 69 ff.
Nakhla, valley southwest of Mecca,

seat of Al-Uzza worship, 236.
Nana, a name of Ishtar at Kish,

181 ff. ; etymology of, 182 ; goddess

of Erim, 185; a Semitic

186 ff.; of Khallabi, 199, 260 ff.;

applied to Ishtar at Erech, 256 ff.

Naphtali, a Hebrew clan, 273.

Naram-Sin, king of Agade, 153 ; date
of, 166, 159; called " god," 168, 199.

Nasamones, a branch of Berber
Hamites, 117.

Nasr, Arabic vulture god, 36.

Nebuchadnezzar I, king of Babylon,
cir. 1130 B.C., 154, 219.

Nebuchadnezzar II, king of Babylon,
604-562 B.C., 154; repaired temple
of Ishtar at Erech, 257.

Nejd, a region in central Arabia, 4.

Nergal, god of Kutu or Kutha, 215;
later, god of underworld, 216; ori-
gin of, 216 ff . ; a Sumerian god, 217 ;
worship in Palestine, Phoenicia, and
Athens, 217, n. 2.

Nestle, E., 284.

Neubaur, A., 230.

Neumahr, Melchior, 18.

Nidaba, the grain god of Gishgalla,218.

Nina, goddess of the city Nina, 185;
a Semitic goddess, 187 ff. ; cult of,
188 ff. ; picture of, 189; daughter of
Ea, 195.

NinS., one of the districts of Shirpurla,
185 ; a Semitic settlement, 192 ; col-
ony of Eridu, 198, 200.

Ninagidkhadu, a name of Ishtar of
Erech, 256.

Nineveh, capital city of Assyria, 261 ;
name written by same ideogram as
Nina, 187, 261 ; a Semitic town, 222 ;
Ishtar goddess of, 261.

Ningirsu, god of Girsu, 185 ; husband
of Ban, 191; origin of, 191 ff.;
"king" and "warrior" of Enlil,
191, n. 7, 196, 202; connected with
Tammuz, 193; developed Ishtar,
193 ff. ; became a sun god, 194.

Ningishzida, deity developed from
Bau-Ishtar, 190; coordinate with
Tammuz, 264.

Ninib, another name for Ningirsu, q.v.

Ninkharsag, originally an Ishtar, 186.

Ninlil, spouse of Enlil, 203.

Ninmar, a goddess, daughter of Nin^,
188.

Nippur, one of the oldest Babylonian
cities, 162; first held hegemony in
Babylonia, 163; antiquity of, 202;
Sumerian foundation," 204; some
Semitic elements, 204 ff.

GENEBAX INDEX

833

Noldeke, Theodor, 3, 6, 22, 44, 53, 88,
227, 278.

Nowack, Wilhelm, 85, 93, 99, 115, 238,
296.

Numerals, Hamitic and Semitic alike,
9 ; the decimal system, 170 ; Sume-
rian system sexagesimal, ibid.

Nusku, Assyrian fire god, origin un-
certain, 232, u. 2.

Cannes, Berossos's name for Ea, 196.

Oasis, centre of Arabian life, 33, 39;
Arabian life bound up in, 71; proto-
type of Eden, 96, n. ; sacred tracts,
112; In North Africa, 117; signifi-
cance of, in Semitic religion, 179 ff.

Odyssey, 249.

Oehler, 6. F., 283.

Oman, women of, 44, 57.

Oppert, Jules, 158, 210, 225.

Osiris, water god of Egypt, 116.

Ovid, 242.

Ox, symbol of Sin, significance of,
201.

Palestine, physical features of, 144;
produce of, 145; conquest of, by
Semites, 146.

Palgrave, William Gifford, 6, 17, 44,
47, 57, 74, 75, 77.

Palm, date, cultivated in Arabian
oases, 33, 39; extent in prehistoric
time, 75 ; in Arabia, 75, n. 4 ; names
of, in Semitic languages, 76 ; primi-
tive name, ibid. ; importance to
Arabian and primitive Semitic life,
77 ff. ; cultivation of, 78 ; sexes of,
recognized, ibid. ; gives name to a
sacred place, 79, 126 fE.; connected
by Semites with knowledge of hu-
man sexuality, 79, 91; sacred in
Arabia, 88; Israel, 89 ff., Bab., 90fE.,
159 ff. ; artificial fertilization of, 91,
161 ff. ; significance of, in Gen.,

92 ff. ; tree of knowledge and life,

93 ff.; Arabic poem on, 95; etymol-
ogy of Ishtar connected with, 104 if . ;
in Egypt, 117, 118 ; at basis of Ham-
ito-Semitic culture, 119; picture of,
in Babylonian writing, 160; neces-
sity made oasis countries earliest
seat of its culture, 162 ; a palm or-
chard in Babylonia, ibid. ; palm in
Semitic religion, 179 ff. ; sacred
palm at Eridu, 197 ; at Nakhla, 236 ;
among the Kenites, 285; at Sinai,

ibid. ,' a totem in a Judahite clan,
286 ; connection with Yahwe, 286 ff.

Palmyra, an oasis 150 miles northeast
of Damascus, 232 ; Al-Lat and Sha-
mash worshipped there, 234 ff.

Paphos, seat of worship of Ashtart in
Cyprus, 249 ff .

Passover, a festival, 108 ff. ; nomadic
festival, 109; festival of Yahwe,
281 ff.

Paulitschke, Philipp, 21, 55.

Pausanias, 250, 253, 266, 267, 313, 314,
315, 316, 317.

Payne, Edward John, 14, 39, 157, 162,
172, 173, 175, 176, 177.

Peake, Arthur S., 141.

Peiser, F. E., 46, 53, 219, 311.

Perrot and Chipiez, 173.

Persians, Semitic influence on, 310 ff.

Peru, civilization and religion of,
175 ff.

Peschel, Oscar, 12, 13, 16.

Petermann, A., 75, 79, 98, 161, 171.

Peters, John P., 156, 199, 204, 292.

Petra, capital of Edom, seat of the
worship of Al-Lat, 233 ff.

Petrie, W. M. Flinders, 274.

Philistines, origin of .uncertain, 231,n. 6.

Philo of Byblos (Gebal), 37, 150, 231.

Philostorgius, 99.

Phoenicia, beginnings of its religion,
149; goddesses in, 243 ff.

Piepenbring, Charles, 281, 284, 291.

Pietschmann, Richard, 93, 112, 245.

Pinches, T. G., 184.

Pindar, 314.

Pithecanthropus erectus, 13.

Pliny, 239.

Polyandry, 30, 39 ff. ; not universal,
40 ff. ; theory of, 52 ff. ; did Semites
practise? 59 ff.; in India,, ibid. ; in
Thibet, ibid. ; among Britons, 60;
in many parts of the world, ibid.;
causes of, ibid.; Semitic, 62 ff.;
combined with polygamy, 63; Thi-
betan displaces Nair, 65 ff., 69 ff. ;
began before Semitic dispersion, 68 ;
imposed restraints on men, 72 ; not
found in lowest social develop-
ments, 72 ff. ; among Hamites, 74,
115 ff. ; among Nasamones, 117 ;
effects of Semitic, 322.

Polybius, 253, 266.

Polygamy, 40 ff. ; mingled with poly-
andry, 61 ff. ; among the rich in
agricultural countries, 171 ff.

334

GENERAL INDEX

Prescott, William, 175, 176, 177.
Prestwich, Joseph, 156.
Price, Ira M., 164, 185, 186, 188.
Pronouns of Hamitic and Semitic

alike, 9.
Ptolemy, Claudius, 32.

Qa'aba, sanctuary of Allah at Mecca,
132 ff . ; originally a heathen shrine,
ibid. ; Al-Uzza worshipped at, 235 ff.

Qazwini, an Arabic writer, 78.

Quatrefages de Bre'au, Jean A. de, 13,
14.

Kaces, origin of white, 15 ff. ; divi-
sions of, 16 ; Mediterranean, 18.

Eadau, Hugo, 157, 158, 160, 161, 163,
166, 167, 168, 169, 180, 181, 182, 184,
186, 187, 188, 189, 192, 197, 198, 199,
200, 202, 205, 211, 213, 214, 217, 220,
225, 229, 251, 256.

Bamman, temple of, in Ashur, 222;
in prehistoric time, 226; extent of
his worship, 224; worshipped by
Anu-banini, etc., 225; member of
second triad, ibid.; name means
"thunderer," 226; a god of fertil-
ity, 228 ; a transformed Ishtar, 229 ;
an ox, 228, u. 7.

Eamsay, William M., 312, 313, 314.

Eatzel, Friedrich, 12, 24, 25.

Eawlinson, George, 156.

Eeclus, Elie, 40, 59, 61, 124, 125.

Eeinisch, Leo, 48, 55, 100.

Eeisner, George A., 31, 106, 157, 158,
165, 205, 228, 246.

Eenan, Ernst, 321.

Eeubenites, 271; in Egypt, 272.

E(!ville, Albert, 148, 173, 174, 177.

Ehea, a Phrygian goddess, same as
Attis, 313.

Ehodes, Ashtart worship in, 252.

Eibeiro, a geologist, 14.

Richter, Max Ohnefalse, 87, 252.

Eidpath, John Clark, 13, 15, 25.

Eipley, William Z., 7, 16, 17, 18, 19, 20.

Eitter, Carl, 78.

Robertson, T., 275,292.

Eobinson, George L., 152.

Eogers, Robert W., 156, 163, 192, 199,
202, 217, 220.

Rome, mother goddess in, 178; Ash-
tart worship in, 253.

Eouvier, Jules, 230.

Sabsea, economic condition of, 124;

kingdom of, succeeded Ma'in, ibid. ;
probably had a precursor in north
Arabia, 125.

Sacy, S. de, 79.

Sahara, once submerged, 18.

Saho language, 10.

Sale, George, 39.

Salkhad, a seat of the worship of Al-
Lat, 233.

Samura, name of a tree at Nakhla,
236.

Sargon,kingof Agade,153; conquered
Amorites, 146; father of Naram-Sin,
155; deified, 168; worshipped Sha-
mash and Ishtar, 213.

Sarpanit, goddess of Babylon, spouse
of Marduk, 207 ; name derived from
sun, 209; a reflection of Marduk,
210.

Sarzec, Ernest de, 160, 180, 183, 184,
186, 187, 188, 189, 190, 191, 192, 193,
198, 207, 210, 211, 214, 218, 220, 260,
261.

Sayce, A. H., 4, 24, 102, 106, 118, 196,
197, 209, 211, 228, 232, 283.

Schaff, Philip, 317.

Soheil, v., 156, 181, 183, 184, 196, 201,
310.

Schmidt, Nathaniel, 8, 25, 272.

Schrader, Eb., 5, 28, 29, 43, 90, 102,
106, 146, 168, 214, 229, 230, 284, 285.

Schultz, H., 284.

Schiirer, Emile, 318.

Schwally, Friedrich, 95, 121.

Semiramis in myth of Dekerto-Atar-
gatis, 243.

Semites, theories of cradle land, 1 ff. ;
relation of language to Hamitic,
9 ff . ; said to belong to black race, 16 ;
evidence insecure, 17 ; one stock with
Hamites, 20 ; home in North Africa,
21; why entered Arabia, 27, 119; ra-
cial characteristics, 28 ; clans, 30 ff . ;
totemistic, 35 ff. ; economic purpose
of clans, 38 ff. ; family, 39 ff. ; sexual
propensities of, 41 ff. ; beena mar-
riage of, 55 ; polyajujry of, 61 ff. ;
passed savagery when separated
from Hamites, 73 ; knew date-palm,
75, n. 4; early religious conceptions,
81; sanctuaries, sacrifices, etc., 106;
gathered in oases for date harvest,
111 ; driven to Arabia by crowding
of African oases, 119; mixture of
religion of, with foreign, 147 ff.; in-
fluenced Babylonian syllabary, 168,

GENEEAI, INDEX

335

u. 1 ; goddesses of, 179 ff. ; at Shir-
purla, Lulubi, Kish, Guti, and Agade,
186, n. 1 ; influence on civilization,
ch. viii ; both evil and good, 309 fE. ;
good predominates, 322.

Sennacherib, first to mention Ishtar
of Arbela, 262.

Sergi, G., 8, 16, 17, 18, 20, 26, 117, 270,
312, 313.

Serpent, nature of, in Eden, 93, 96 fE.

Sex, perception of, represented as fruit
of tree of knowledge, 94 ; what Sem-
ites attributed to knowledge of,
102 ; connection with moral advance-
ment, 107; and religious feeling,
307ff.,321fE.

Shalraeneser II, king of Assyria, 154;
called Hadad, " Bir," 227.

Shamash, god of Agade and Larsa,
212; worshipped at Agade by Sar-
gon, 213 ; antiquity of, 213 ff. ; orig-
inally a goddess, probably Sumerian,
214.

Shamashshumukin, Semitic idiom in
Sumerian prayer of, 165.

Shams, sun goddess, a survival of the
mother goddess in south Arabia,
129 ff.

Shamsu-iluna, Babylonian king, built
a fortress to Ramman, 225.

Sharastani, Mohammed, 99.

Shidlamtsea, an epithet of Nergal, 215.

Shirpurla, temple taxes of, 158 ; dates
received at, 159 ; city of old Baby-
lonia, 162 ; held hegemony at dawn
of history, 163 ; religion of, 184 ff . ;
districts of, 185 ; three districts Se-
mitic, 185; language spoken at,
186 ff. ; name written without de-
terminative, 260 ff.

Sicily, Ashtart worship in, 252 ff.

Sidon, religion of, 150 ff. ; Ashtart
worship in, 243 ff.

Simeon, a tribe, 271.

Sin, a god, father of Mana, 199; god
of Ur, ibid. ; a transformed Ishtar,
200; identified with moon, 202; son
of Enlil, ibid. ; member of the second
triad, ibid. ; worshipped by the Guti,
217.

Sin-gamil, a king of Ur who repaired
temple of Nergal, 216.

Sippar, another name of Agade, 212;
Khallabi near, 260.

Skipwith, Mr., 283.

Smend, Rudolf, 140, 284.

Smith, George, 92, 197, 239, 241, 261,
262.

Smith, George Adam, 121, 139, 145,
156, 230, 238.

Smith, Henry Preserved, 89, 132, 275.

Smith.I. G., 317.

Smith, W. Robertson, 7, 29, 30, 31, 32,
35, 36, 39, 41, 51, 52, 54, 55, 57, 59,
63, 64, 65, 66, 68, 70, 76, 81, 86,87, 88,
89, 92, 100, 101, 104, 105, 106, 107, 108,
109, 112, 114, 123, 127, 132, 133, 136,
138, 148, 152, 234, 235, 237, 251, 264,
255, 265, 267, 279, 281, 283, 287, 289,
291, 312.

Social organization, ch. ii; effect on
religious conceptions, 82; trans-
formed in Arabia by time of Mo-
hammed, 123.

Socin, Albert, 140.

Somali language, 10.

Somalia, Somaliland, 8, 20, 24, 25, 48,
56.

Sozomen, 99.

Spencer, Herbert, 40, 63, 69, 60, 63, 66,
72.

Spiegelberg, Wilhelm, 283.

Sprenger, A., 4, 24, 44, 138.

Springs, sacred among the Semites,
92 ff.

Stade, B., 95, 121, 147, 238, 270, 275,
279, 284, 296.

Starbuck, J. M., 107.

Starcke, C. N., 40, 49, 51, 53, 59, 60,
66, 67.

Steindorf, G., 10, 247.

Steurnagle, Carl, 251.

Strabo, 42, 64, 101, 115, 123, 210, 260,
253, 313, 314, 315.

Strack, Hermann L., 275.

Strassmaier, J. N., 46, 106, 226, 260.

Stumme, H., 19.

Suess, Ed., 13, 18.

Sumerians, problem of, 164 ff. ; argu-
ments of the Halevy school, 164 ff. ;
counter arguments, 165 ff, ; solu-
tion, 167 ff. ; inventors of cuneiform
writing, 167 ; analogy of El-Amarna
letters, 167 ff. ; religious argument,
168; pictorial evidence, 170; nu-
merical system, 170 ff. ; goddesses
of, 180 ff. ; Sumir said to be original
form of Girsu, 192, n. 1; Sumerian
kingdom at Nippur, 204.

Sumula-ilu, Babylonian king, witness
to the early worship of Marduk,
209.

336

GENEBAL INDEX

Surippak, a city mentioned in Baby-
lonian deluge story, 259.
Suti or Guti, 37 ; their religion, 217 ff.

Tabari, At-, 88.

Tabernacles, feast of, 108; descended
from old Semitic date festival, 111,
288 ff.

Tacitus, 250, 251, 252, 266.

Taif, south of Mecca, seat of worship
of Al-Lat, 233.

Talab Riyam, a south Arabian god,
developed from Athtar, 129.

Tallquist, K. L., 158, 159.

Tamesheq language, 10.

Tammuz, general features of his
worship, 85 ff. ; various relations
to Ishtar, 85; connected with vege-
tation, 85 ff., 86 ff., 112 ff. ; origin of
wailing for, 86, 92 ff., 114 ff.; festi-
val of, originally a fast before date
harvest, 112, 289; connected with
Bau-Ishtar, 190; with Ningirsu, 193 ;
bewailed at Erech, 258; survivals
of, 263 ff . ; nature of, 264 ff . ; called
in Phoenicia Eshmun and Adon,
265 ff. ; connected with caravan
clans, 268; in Yahwe worship, 289 ff.

Tanith, name of Semitic goddess in
North Africa, 148, 150, 263; mean-
ing and etymology of name, 253,
n. 6; mother goddess, 254; same as
Dido, 254 ff.; "face of Baal," 150,
254.

Tarkhu, Hittite god, held by Jensen
to be original of Atargatis, 240.

Tashmit, consort and reflection of
Nabu, 212.

Tell-Ibrahim, modern name of Kutha,
215.

Telloh, modern name of Shirpurla,
183.

Tertullian, 255.

Teutonic pantheon, 178.

Thatcher, Oliver, 318.

Theodulus, son of Nilus, 237.

Thibetan polyandry, 61, 63 ff.; in
Yemen, ibid. ; relation to Nair,
69 ff.

Thomas, Richard H., 82.

Thresliold, sacredness of, 101 ff.

Tiele, C. P., 163, 168, 222, 223, 275.

Tiglath-pileser I, king of Assyria, 154 ;
called Ramman, god of " west coun-
try," 226; mentioned Aramaeans,
271.

Tigris, overflow of, 156.

Totem, a clan shibboleth, 33 ff. ; proofs
of, 35 ff. ; among Hebrews, 36 ff. ;
neighboring tribes, ibid. ; break up
of, 63 ; trees as, 87 ; at Erech, 268.

Toy, C. H., 36, 86, 91, 95, 113, 282, 295.

Tree worship, 87 ff. ; in Abyssinia, 89 ;
in Palestine, 89 ff. ; significance of,
in Eden, 93-97 ; at Eridu, 197.

Trumbull, H. Clay, 101, 282.

Tylor, E. B., 91.

Tyre, religion of, 149 ff. ; worship of
Ashtart in, 244; wealth of, 300.

Umu, a name of Ishtar at Erech, 256.

Ur, city of Babylonia, 162 ; its dynasty,
157; its temple taxes, 158; its god.
Sin, 199; held a prehistoric hege-
mony, 199, 256.

Ur-Bau, Patesi of Shirpurla, 153; wor-
shipped Nana of Erim, 186; Nin3,,
188; Ea, 198; Dumuzizuab, 211;
meaning of his name, 251, n. 2.

Ur-Enlil, an old Babylonian king,
204, n. 1.

Ur-Gur, king of Ur, 153, 202 ; repaired
temple of Shamash at Larsa, 212,
256.

Urkagina, king of Ur, 153 ; worshipped
Ban, 189, 194.

Ur-Nina, king of Shirpurla, 153 ; wor-
shipped Nin&, 188 ; Ban, 189; mean-
ing of his name, 251, n. 2.

Ur-Ninib, king of Isin, 256.

Uruazagga, one of the districts of
Shirpurla, 185; a Semitic settle-
ment, 192.

Uruk, ancient name of Erech, 256.

Urzaguddu, king of Kish, 203.

Venus, Latin name of Ashtart, 249.

Virgil, 253, 255.

Vlock, W., 3, 28.

Vogue, Comte de, 232, 233, 234, 239.

Wadd, a Minsean god, 1,S0 ff. ; name
derived from root " to love,"
131 ff. ; developed from mother
goddess, ibid. ; consort of Athirat,
247.

Waitz, Theodor, 60.

Wallace, Alfred Russell, 15, 18, 26, 74.

"Ward, Wm. Hayes, 184, 189, 194, 210.

Weber, Ferdinand, 281.

Weber, Otto, 124, 125, 130, 272.

Weeks, feast of, 108; not primitive.

GENERAL INDEX

337

112 ff . ; adapted from Tammuz wail-
ing preceding date festival, 113 ff.

Weissbaeh, F. H., lt>i, 168.

Wellhausen, Julius, 36, fl3, 63, 6S, 69,
72, 79, 88, 89, 101, 108, 109, 111, 112,
133, 134, 136, 147, 227, 233, 234, 235,
236, 237, 238, 270, 280, 281, 283, 292,
296.

Wells, sacred among Semites. 92 ff.

Wellsted, J. R., 57, 74, 75, 77, 78, 111,
285.

Westerraarck, Edward, 40, 41, 49, 51,
53, 60, 66, 67.

Wheat, indigenous to Babylonia, and
one of the first grains there culti-
vated, 157 ; references to it in lit-
erature, ibid., n. 2.

White, H. A,, 243.

White races, origin of, 15 ff. ; divisions
of, 16.

Whitney, J. D., 14.

Wiedemann, A., 8, 20, 26.

Wife, residence of, during marriage,
50 ff. ; residence of husband in her
tent or tribe, 54 if. ; subject to hus-
band, 123.

Wildeboer, G., 272, 275.

Wilken, G. A., 39, 47, 48, 61, 63.

Wilkinson, John Gardner, 296.

Winckler, Hugo, 31, 64, 106, 109, 111,
113, 130, 162, 163, 181, 198, 199, 203,
209, 217, 227, 255, 270, 272, 274, 279,
286, 289, 296.

Winstanley, W., 49, 137.

Women, Semitic, beauty fades early,
42 ; of Oman, 44 ff . ; lived in homes
of brothers or uncles, 45 ff., 49 ff. ;
residence of during marriage, 50 ff . ;
exalted position of, 53; kept chil-
dren with them, 54; not creatures
of man, 56 ff . ; in Babylonia, 57 ff. ;
position modified, 58 ff. ; scarcity
of, 62; liberty limited in Thibetan

polyandry, 70 ; desert women lower

order, 71; sacrifices of, in Adonis

worship, 246.
Woodhouse, F. C, 317.
Wright, William, 5, 10, 24, 28, 29, 312.
Wustenfeld, Ferdinand, 44, 235.
Wylde, Augustus B., 25, 49, 99, 136,

138.

Yaguth, Arabic lion god, 35.

Yahu-bldi, king of Hamath, 284, 285.

Yahumelek, king of Gebal, 122, 244.

Yaliwe, origin of, ch. vii; a Kenite
god, 272; supporters of this view,
275 ; name revealed at Horeb, 276 ;
unknown to fathers, ibid. ; home at
Sinai, 277 ; storm god theory of
Yahwe, 279 ; a transformed mother
goddess, 280 ff., 287; etymology of
name, 282 ff. ; connected with palm,
286 ff.; Yahwe's passover, 287 ff. ;
connected with Ashtart and Tam-
muz, 289 ff. ; spiritual possibilities
of covenant with, 291, 301 ff. ; be-
came a Baal, 297 ff. ; genesis of
spiritual conception of, 302 ff. ; con-
nection of law with, 304 ff.

Yaqut (Jacut), 44, 63, 76, 234.

Ya'uq, Arabic horse god, 36.

Yeha, town in Abyssinia, 135 ff.

Yemen, economic condition of, 124.

Zebulon, an Israelitish elan, 271.
Zemzem.the sacred spring at Mecca,

133, 235, 236.
Zimmern, Heinrieh, 10, 87, 103, 140,

165, 200, 214, 215, 260, 274.
Zira-bani-ti, a folk etymology of the

name Sarpanit, 209, n. 8.
Zohair, an Arabian poet, 52.
Zwemer, S. M., 48, 77, 78, 91, 95, 111,

124, 125, 132, 134, 159, 161.

INDEX OF SCRIPTURE REFERENCES

Genesis pasx

ii 93, 96

ii, 8 96

il, 9 95

ii, 14 96

ii, 15 96

Hi 23, 89, 93, 96

iii, 3 95

iii, 9 95

iii, 22 95

xii,4 271

xii, 18 89

xiv, 5 142

xiv, 13 238

xvii 280

xvii, 10-12 99

xviii 140

xviii, 1 89

xix 140

xxiv : 271, 278

xsdv, 2 281

xxiv, 9 281

XXT, 12 ff. 32

xxviii, 1-xxxii, 1 . . . . 271, 278

xxviii, 22 290

xxxi, 43 64

xxxiv 100, 281

XXXV, 8 89

xxxvi 32

xxxviii 90, 286

xlvi, 17 31

xlvii, 29 281

xlviii, 5, 6 52

xlix, 5 ff 272

Exodus

ii, 16ff 280

iii, 1 280

iii, 13 276

iii, 14 283

iii, 15 276, 283

iv, 24, 25 . . . . 99,100,280,281

PAOK

vi, 2 276

xii, 48 99, 280

xiii 279

xiv 279

XV, 27 90, 285

xviii, 12 ff 272, 276

xix 279

XX 292

xxi, 12-14 138

xxiii, 16 288

xxxii, 2 296

xxxiv 109,292

xxxiv, 18-20 110

xxxiv, 22 ff 288

LEViTicns

xvi 305

xxiii, 34 288

xxiii, 40-43 288

XXV, 49 51

Numbers

V, 11-21 305

X, 29 ff 277

xxi, 27-30 142

xxi, 29 141

xxvi, 45 31

xxxli, 12 251

Deuteeonomt

i, 39 94

V 292

vii, 5 290

vii,13 105,282

xxiii, 17, 18 251

xxiv, 1-3 45

xxvi, 5 . . . . 271

xxviii, 4, 18 282

xxxii, 17 87

xxxiii, 2 277

xxxiii, 29 249

xxxiv, 3 90, 285

.339

340

INDEX OF SCRIPTURE REFERENCES

Joshua paoe

V, 3, 9 280

xiii, 21 238

xiii,31 142

XT, 32 227, 228

xxiv, 14 276

JCDQES

i, 16 90, 277, 285, 288

ii, 13 246

iii, 7 247

iii, 13 90

iv, 5 89

iy, 11 277

iv, 17 ff 277

■<^ 57

V, 4ff 277

T, 24 ff 277

Ti, 11 89

vii, 25 36

X, 6 148,246

xi,40 289

XTi, 23 230

XX, 33 90

XX, 47 277

xxi, 13 227

1 Samtjei.

i 287

ii 288

vii, 4 246

XV, 6 277

xxvi, 19 297

XXX, 26 ff 278

XXX, 29 278

xxxi, 4 280

xxxi ,9 242

2 Samubi.

i,20 280

V, 2 ff 230

viii, 3 227

1 Kings

i 138

ii 138

viii, 9 295

viii, 10, 11 279

viii, 21 295

ix, 8 286

X, 1 ff 124

xi, 7, 33 141

xii, 28 37, 298

FAGB

XV, 18-20 226

xviii, 19 247

xix 277

XX 226

xxi 300

2 Kings

iii, 4 140

V, 17 227, 299

v, 18 226

vi, 24 226

X, 15 277

xvii, 24-34 139, 147, 149, 203, 217, 297

xvii, 27, 28 299

xxli, 13 304

xxiii, 4 247, 291

xxiii, 13 151

xxiii, 14 290, 291

1 Chronicles

ii,55 277,278

vii, 31 31

X, 4 280

Job

xxxvii, 4 279

xxxviii, 1 279

PSALIUS

xviii 279

xviii, 10 91

xxix, 3 ff 279

Ixviii, 5 [4] 277

Ixviii, 33 279

civ, 13, 14 279

cxlvii, 8, 16-18 279

Isaiah

i, 13-15 302, 304

V, 1-7 302

xix, 1 279

1, 1 45, 302

Ixv, 11 249

Ixvi, 17 36

Jeremiah

ii,28 148

iii, 1 ff 302

iii, 4 68

vii, 18 246

xi, 13 148

XXXV . 277

xlviii, 7, 13, 46 141

INDEX OF SCRIPTURE REFERENCES

341

EZEKIEI. p^GE

i 279

vili, 1 113

viii, 10 36

viii, 14 85, 246

XX 302

xxii, 11 52

xxvii 300

xxviii 300

xli, 18 90, 286

HOSEA

ii 302

ii, 5 151,297

ii, 8 298

ii, 12 151

iii, 4 290

iv, 13 89

iv, 15 298

vi, 6 304

ix, 10 144

ix, 15 298

xi, 1 ft 306

xii, 11 298

Amos

iii, 2, 3 302

V, 4fE 298

T, 21 ft 298, 304

V, 24, 25 302

vi, 13 238

vii, 14 145

Habaekue

iii 279

iii, 1 277

ZeCHAEIAH piOE

xii, 11 227

LUEE

XV, 16 145

John

iv, 24 303

xvii, 23 107

Acts

xix 314

Romans

ii, 28fE 281

iii, 30 280

1 COKDfTHIANS

V, vi 317

X, 20 87

xiii 302

2 Petkb

i, 21 307

1 John

i, 5 302

iv, 8, 16 302

BBTEIiATION

xxi 96

xxii, 1, 2 96

i,5

REFERENCES TO APOCRYPHA

TOBIT

37

Epistle of Jeremiah
42, 43 42, 210

1 Maccabees

v,43. .
X, 83, 84
xi, 4 . .

238
230
230

2 Maccabees
xii, 26 238

Ethiopic Enoch

X, 10 96

xxiv 89,96,28li

XXV 96

Psalter of Solomon
xvii . 96

342

REFERENCES TO QTJR'AN

REFERENCES TO QUR'AN

PAGB

ii, 140 132

iv 131

iv, 23 65

iv, 26 64

iv, 29 123

V, 2, 3 132

V, 96-98 132

Ti, 100 36

PAOK

xix, 23 89

xix, 25 «9

xxxiii, 48 ^i

liii 131

liii, 19 235

Ixv, 1-6 46

IxT, 6 50

cxii 131

History, Prophecy, and the
Monuments.

JAMES FREDERICK McCURDY, Ph.D., LL.D.,

Professor in Oriental Languages in University College, Toronto,

IN THREE VOLUMES.

8vo. Cloth. $3.00, net, each.

"We must characterize this work as of the very highest merit,
written in the most lucid style, and an admirable statement of the
present state of our knowledge." — Queen's Quarterly.

"The reader will close the volumes with a wonderfully richer con-
ception of the Old Testament story."

— Cincinnati Commercial Gazette.

"A work of vast erudition, at the same time readable and interest-
ing. . . . The author has given us what must assuredly be considered
the standard work on that branch of Biblical history to which he has
devoted himself." — Church Almanack.

"A work of which American scholarship has just reason to feel
proud. It is a notable contribution to an important subject. ... It
is to be heartily recommended to the general public as a very useful
compendium. For Bible readers and Bible students alike, it is an
invaluable guide." — New World.

THE MACMILLAN COMPANY,

6(5 FIFTH AVENUE, NEW YORK.

I I i.'.hl.) I 'J.li '.).'. ','.'■ 'k'.^'l'Lh'.','L^'i^'l^'l'l'L'l\'l'l^^'^'l'l'L'i'l'i\\^'^'l'iK^'i'l'L'l\'l'^'^'l'l'^^'l'^